AI-driven clinical recommendations may aid physician decision making to improve quality of care

Embargoed for release until 10:00 a.m. ET on Friday 4 April 2025
Embargoed Content from the Annals of Internal Medicine Breaking News Scientific Plenary at Internal Medicine 2025

Annals of Internal Medicine Tip Sheet     

@Annalsofim    
Below please find summaries of new articles that will be published in the next issue of Annals of Internal Medicine. The summaries are not intended to substitute for the full articles as a source of information. This information is under strict embargo and by taking it into possession, media representatives are committing to the terms of the embargo not only on their own behalf, but also on behalf of the organization they represent.    
----------------------------    

1. AI-driven clinical recommendations may aid physician decision making to improve quality of care

Overall, in real life clinical settings, AI recommendations were rated as superior to physicians with less risk for potential harm

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-24-03283

Editorial: https://www.acpjournals.org/doi/10.7326/ANNALS-25-00472   

URL goes live when the embargo lifts    

A study comparing artificial intelligence (AI)-driven clinical recommendations to physician decision-making for common acute complaints in a virtual primary care setting found that AI support has the potential to improve quality of care when implemented responsibly. In the study, AI recommendations were largely rated as superior to physicians by expert adjudicators. Trained on a massive amount of high-quality, real-world clinical data, which is far beyond what any physician would see in their lifetime, AI uses modeling to calculate real probabilities for diagnosis and referrals and then derives treatment plans from these probabilities based on the latest medical guidelines and protocols. Combined with human physician discernment, this approach could improve patient care and outcomes. The findings will be presented at the breaking news scientific plenary session "New in Annals of Internal Medicine: Hear it First from the Authors" held at the New Orleans Ernest N. Morial Convention Center during the American College of Physician’s (ACP) Internal Medicine Meeting 2025. The paper also is published in Annals of Internal Medicine.

Researchers from Tel Aviv University, Cedars-Sinai Medical Center, and K Health, Inc. compared initial AI-driven clinical recommendations to final physician decision-making at Cedars-Sinai Connect, a virtual primary care clinic run by Cedars-Sinai, leveraging K Health Clinical AI platform. To evaluate AI versus physician performance, four expert adjudicators reviewed 461 real life primary care visits involving acute urinary, vaginal, respiratory, eye, and dental chief complaints, conditions for which AI has shown high diagnostic accuracy, and used a checklist to rate both the AI and physician recommendations. Patients described their symptoms through an AI-guided medical intake with their medical records automatically synced from the EHR, allowing physicians to review the AI-generated diagnosis, treatment recommendation, and any suggested prescriptions, labs, or referrals before the virtual visit where doctors would make the final clinical decision. Physicians could access the initial AI decisions but the researchers were unable to determine whether they did before making their final care decisions. Overall, the expert adjudicators rated the AI's recommendations higher than physicians’ decisions. In about 2/3 of cases, doctors made the exact same clinical decisions as the AI. In the remaining 1/3 of cases, AI's recommendations were rated as superior twice as often as they were inferior. AI received only about half as many "potentially harmful" ratings compared to physicians.

According to the researchers, these findings demonstrate the strengths of AI and how it complements human practice. AI is designed to follow medical guidelines much more strictly and can pick up on subtle details or medical record notes that doctors may miss. However, doctors have a clear advantage in clarifying symptoms during the visit. Where AI and good doctors align is in their level of caution. If AI is not confident, it will not make a recommendation, which is what happened in about 20% of cases.

The author of an accompanying editorial from Tufts University School of Medicine points to the study strengths while offering notes of caution about incorporating AI into medical practice. A common-sense approach to using AI may include using it to confirm physician diagnoses and therapeutic recommendations rather than making them. If discrepancies exist, then physicians should seek to determine why, which can be the seeds for improving the AI tool. The author emphasizes that to enhance clinical relevance, experienced clinicians should be involved in AI program development.

Media contacts: For an embargoed PDF, please contact Angela Collom at acollom@acponline.org. To speak with an author, please coordinate an interview with an author, please email Lauren Hobson at lhobson@moxiegrouppr.com or Michal Kaplan-Nadel at michal@moxiegrouppr.com.

----------------------------   

2. Switching to tirzepatide rather than increasing dulaglutide dose leads to better glucose control and more weight loss in patients with inadequately controlled diabetes

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-24-03849    

URL goes live when the embargo lifts     

For patients with inadequately controlled type 2 diabetes, switching to tirzepatide improves glucose control and increases weight loss more so than maximizing dulaglutide dose. This is important because delays in attaining at-goal glucose control can increase the risk and progression of diabetes-related complications like kidney disease and eye disease. The findings will be presented at the breaking news scientific plenary session "New in Annals of Internal Medicine: Hear it First from the Authors" held at the New Orleans Ernest N. Morial Convention Center during the American College of Physician’s (ACP) Internal Medicine Meeting 2025. The paper also is published in Annals of Internal Medicine.

The SURPASS-SWITCH (A Phase 4, Randomized, Open-Label, Active-Controlled Study to Investigate the Efficacy and Safety of Switching from Weekly Dulaglutide to Weekly Tirzepatide in Adults with Type 2 Diabetes) trial was a multicenter study conducted in five countries designed to address a common clinical experience for physicians prescribing a GLP-1 RA for the treatment of type 2 diabetes. In the study, sponsored by Eli Lilly and Company and led by Endeavor Health system (Chicagoland, Ill.), researchers across nearly 40 study sites, researchers sought to determine the safety and effectiveness of escalation of dulaglutide dose versus switching to tirzepatide in patients with inadequately controlled type 2 diabetes. Participants were randomly assigned to either continue on dulaglutide at a higher dose or switch to tirzepatide. Medication doses were gradually increased to 4.5 mg of dulaglutide or 15 mg of tirzepatide, or the highest dose the participant could tolerate. The primary end point was change from baseline in HbA1c at week 40. The key secondary end point was change from baseline in weight at week 40. 

The researchers found that switching to tirzepatide resulted in greater HbA1c reduction than escalating dulaglutide dosage, with 21.3% receiving tirzepatide compared with 2.4% receiving dulaglutide achieving an HbA1c value less than 5.7%, the threshold for normal blood glucose levels. Participants in the tirzepatide group also had a mean of 6.9 kg greater weight reduction with 58% of participants on tirzepatide (vs 6.8% on dulaglutide) achieving 10% or greater weight reduction, a threshold that has been observed to have disease-modifying effects. Adverse effects were similar in both groups. The study authors note that further research is needed to assess the potential benefit of early glycemic control via early medication switching on longer-term clinical outcomes.

Media contacts: For an embargoed PDF, please contact Angela Collom at acollom@acponline.org. To speak with the presenting author, Liana K. Billings, MD, please email Baylee Pulliam at BPulliam@northshore.org. 

-------------------------------

3. ACP’s Best Practice Advice addresses use of cannabis, cannabinoids for chronic noncancer pain

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-24-03319

URL goes live when the embargo lifts     

The American College of Physicians (ACP) has issued Best Practice Advice for clinicians whose patients are considering or using cannabis or cannabinoids for management of chronic, noncancer pain. The details will be presented at the breaking news scientific plenary session "New in Annals of Internal Medicine: Hear it First from the Authors" held at the New Orleans Ernest N. Morial Convention Center during the American College of Physician’s (ACP) Internal Medicine Meeting 2025. The paper also is published in Annals of Internal Medicine.

ACP’s Best Practice Advice paper is intended to inform clinicians about the evidence regarding the benefits and harms of cannabis or cannabinoids in the management of chronic noncancer pain and to provide advice for clinicians counseling patients seeking to use cannabis or cannabinoids for chronic noncancer pain.

Cannabis use for medicinal purposes has grown among patients with chronic noncancer pain. When referring to any product derived from the plant, many use the term cannabis, but the term has become interchangeable with colloquial terms such as marijuana, weed, and pot. As of 2024, 24 states in the U.S. and the District of Columbia have legalized cannabis for adult recreational and medical use, and it is legal for medical use only in an additional 14 states.

In its Best Practice Advice ACP says clinicians should:

Counsel patients about the benefits and harms of cannabis or cannabinoids when patients are considering whether to start or continue to use cannabis or cannabinoids to manage their chronic noncancer pain.

• Counsel the following subgroups of patients that the harms of cannabis or cannabinoid use for chronic noncancer pain are likely to outweigh the benefits:
  • Young adult and adolescent patients
  • Patients with current or past substance use disorders
  • Patients with serious mental illness
  • Frail patients and those at risk of falling
• Advise against starting or continuing to use cannabis or cannabinoids to manage chronic noncancer pain in patients who are pregnant or breastfeeding or actively trying to conceive.
• Advise patients against the use of inhaled cannabis to manage chronic noncancer pain.

For many patients, evidence suggests that the known harms of cannabis and cannabinoid use outweigh the potentially small degree of benefit to ease chronic noncancer pain. Additionally, cannabis can be addictive, even if being used to manage chronic noncancer pain. It’s also difficult to apply the information from clinical studies to practice in the U.S. because the potency (delta-9 tetrahydrocannabinol or THC content) of products in dispensaries is typically far higher than that used in studies. Another challenge is that in most U.S. states, patients will obtain cannabis for chronic pain through a dispensary with less medical oversight than they would receive for FDA-approved medications. 

Clinicians are best positioned to provide evidence-based information about the benefits and harms most relevant to an individual patients’ needs and comorbidities so that patients can make an informed decision about starting or continuing cannabis or cannabinoid use for chronic noncancer pain.  For most patients, common treatments and analgesic medications should be recommended first given the limited evidence of small benefit and the known harms associated with cannabis and cannabinoid products.   

ACP has also published a position paper where it recommends a public health approach to address the legal, medical, and social complexities of cannabis use.

This Best Practice Advice is based on a review and assessment of scientific work including a living, systematic review on cannabis and cannabinoid treatments for chronic noncancer pain, a series of living systematic reviews, as well as additional evidence from primary studies.

Media contacts: For an embargoed PDF, please contact Angela Collom at acollom@acponline.org. To speak with someone from ACP, please contact Andy Hachadorian at ahachadorian@acponline.org.  

----------------------------