Practice Management
Many patients are now using artificial intelligence (AI) tools with large language models to fill the information gap between the time their test results show up in their patient-facing health records—which could be the same day—and the time their clinicians provide personalized interpretation and next steps. Experts warn that the accuracy of AI chatbot information often depends on how the patients frame their questions and what details they input along the way. But it turns …
Read More
Using artificial intelligence (AI) to generate notes to document patient visits was linked to improvements in clinician burnout, as published in JAMA Network Open. Researchers surveyed 1,430 clinicians from 2 different health systems who used ambient documentation technology. They found physicians at Mass General Brigham reported a 21.2% absolute reduction in burnout with the tools, as the proportion of clinicians reporting burnout decreased from 50.6% to 29.4% at 42 days and from 52.6% to 30.7% …
Read More
When a physician recommends an important health service, patients are more apt to listen, aren’t they? Maybe not. A randomized clinical trial of more than 22,233 patients published in JAMA Network Open tested whether personalized physician videos or infographics could raise the uptake of influenza vaccinations for adults and children. As it turned out, in the 2023-2024 respiratory virus season, vaccination rates were similar across groups: 46.9% with usual care, 48.0% with video, and 47.5% …
Read More
AI-generated reports using large language models (LLMs) seem inevitable in healthcare, especially when it comes to the time-consuming task of writing up patient discharge summaries, a commentary post in JAMA Internal Medicine states. Right now, the responsibility of compiling summaries is one of the many administrative burdens leading to burnout among clinicians, and leveraging LLM tools could reduce that burden if done thoughtfully and with appropriate data integration. In a related article, researchers found no …
Read More
In a study published in JAMA Network Open, researchers developed an artificial intelligence (AI) model and tested its ability to assist in triaging referrals from primary care to specialized care. Evaluated against a reference standard, the AI tool achieved good overall accuracy of 0.716 (95% confidence interval [CI], 0.694-0.737) of referrals with high specificity (0.801, 95%CI, 0.777-0.822) but lower sensitivity (0.542, 95%CI, 0.501-0.582). In other words, the AI model correctly classified 71.6% of referrals while …
Read More
From 2010 to 2024, healthcare data breach incidents in the United States rose significantly, with the total number of reported breaches increasing from 216 to 566—a 162% rise. Ransomware-involved cyberattacks across all Health Insurance Portability and Accountability Act (HIPAA)-covered entities increased from 0% of breaches in 2010 to 31% in 2021. However, by 2024, ransomware attacks more recently trended down to 11% of incidents (61 of the total 566 incidents for the year). Even more …
Read More
A review published last week in Health and Technology found that AI can read and interpret images more effectively and faster than radiologists, suggesting that AI could help make impact on today’s radiologist shortage. Even so, the industry must be mindful of the ethical and legal implications of AI use. To be clear, the authors state that AI can’t replace human radiologists but rather that AI tools can complement their work, offer guidance, and assist …
Read More
Accuracy among healthcare artificial intelligence (AI) tools has grown exponentially in a short amount of time, and the place for these tools in urgent care might be that much more evident now. Research published in the Annals of Internal Medicine recently concluded that AI has the ability to match and to outperform physicians delivering virtual urgent care. In fact, AI agreed with doctors’ diagnosis and management recommendations 56.8% of the time among the 461 virtual …
Read More
The Food and Drug Administration (FDA) posted an alert this month warning that patients who use continuous glucose monitors (CGMs), insulin pumps, automated insulin dosing systems, and other diabetes devices are at risk of not receiving or not hearing audible alerts from their smartphones regarding their devices. A missed alert could possibly lead to severe hypoglycemia, severe hyperglycemia, diabetic ketoacidosis, and death, the agency says. Because apps for diabetes CGMs and other devices allow users …
Read More
An observational analysis published in JAMA Network Open examined the level of trust Americans have when it comes to health systems’ use of artificial intelligence (AI). Researchers conducted a national survey of 2,039 adults, analyzing factors that could influence their trust, including AI knowledge, health literacy, and experiences of discrimination in healthcare, using a 4-point Likert scale. The results indicate that most respondents have low trust in health systems to use AI responsibly (65.8%) or …
Read More