Public health and law enforcement agencies face a growing challenge in combating the fentanyl epidemic: They lack detailed and timely information on the harmful and potentially deadly substances that drug suppliers increasingly mix into the illicit fentanyl supply. 

Now, a new study from the National Institute of Standards and Technology (NIST) reveals how the prevalence of these adulterants changes over time and across different regions of the country.  Alerting first responders to the prevalence of specific adulterants in local supplies of fentanyl may make the difference between life and death.

“Our goal is that if some new additive pops up, say in Philadelphia, we’ll be able to tell public health officials when that additive is likely to show up in Delaware, Maryland or the West Coast,” said NIST chemist Ed Sisco.

Tardigrades, also called water bears, are famous for their ability to survive nearly anything. They protect themselves in extreme environments using proteins that researchers have developed into an efficient and effective strategy to preserve human red blood cells. Publishing in ACS Applied Materials & Interfaces, the researchers report that the technique, which is simpler than the current freezing method involving glycerol, restored red blood cell counts in anemic mice.

As artificial intelligence (AI) becomes increasingly capable of assisting with health care tasks, a new study by researchers at the Icahn School of Medicine at Mount Sinai has found that adding a brief safety reminder reduced potentially harmful choices by AI models in clinical scenarios. The study, published in the September 26 online issue of Communications Medicine [DOI: 10.1038/s43856-026-01933-8], a Nature Journal, also found that AI models can be influenced by the context and instructions surrounding a clinical decision. The findings, based on millions of outputs, suggest that how AI systems are prompted and guided may remain an important consideration in developing safe and reliable clinical applications, even as increasingly capable AI models and agents become better able to understand users’ intent without lengthy or comprehensive prompts.