Emergency medicine doctors someday might rely on consultation from artificial intelligence (AI) programs like ChatGPT to help them quickly and accurately diagnose patients’ ailments. A new study found that ChatGPT performed about as well as human doctors in diagnosing patients, when both are given the same set of clinical information. “In the end, they were pretty comparable,” said senior researcher Steef Kurstjens, a clinical chemist with Jeroen Bosch Hospital in Nijmegen, the Netherlands. “And as they’re pretty comparable, [AI] might be helpful to speed up the process or enhance the amount of diagnoses at the emergency department.” For the study, two-doctor teams and the artificial intelligence program each reviewed physician’s notes and lab tests for 30 patients treated in March 2022 at the emergency department of Jeroen Bosch Hospital. The researchers used a free version of ChatGPT and a subscriber version. The AI tools and the medical teams then compiled lists of the top five potential diagnoses for each patient, based on the information at hand. Because these were past cases, the researchers already knew what the exact diagnosis had been. Doctors had the correct diagnosis within their top-five list 87% of the time. By comparison, the free version of ChatGPT had the correct diagnosis listed 97% of the time, and the subscriber version of the AI program 87% of the time. “It’s a nice… read on > read on >
All Gear:
Web Searches for ‘Self-Managed Abortion’ Rose After Dobbs Decision
When some U.S. states made abortion illegal after the Supreme Court overturned the longstanding Roe v Wade in June 2022, women in those areas increased their searches for self-managed abortions. To come to that conclusion, researchers from the University of California, Irvine (UCI) analyzed Google search results regarding self-abortion. “We found an increased number of searches in states where abortion had recently become illegal, suggesting that a lot of people in these states are trying to learn how to give themselves abortions instead of going to the doctor,” said lead investigator Sean Young. He is a professor in the departments of emergency medicine and informatics. “This has implications for increased emergency department visits for self-managed abortions in these states,” Young said in a university news release. This could cause increased problems for Black women who already face challenges with accessing quality health care. Black women receive abortions at a higher rate than white women, according to the U.S. Centers for Disease Control and Prevention. “Online search results may return unsafe home remedies that could have life-threatening consequences for women who seek to self-manage an abortion,” warned Lidia Flores, from the UCI department of emergency medicine. And Dr. Carrie Chandwani, an associate clinical professor in UCI’s department of emergency medicine, added that “emergency medicine providers need to be prepared for any potential increase in injuries, complications… read on > read on >
Americans Are Worried About AI in the Workplace: Poll
Could an algorithm take your job someday? Concerns about artificial intelligence, or AI, are plaguing U.S. workers, according to a new American Psychological Association poll. Some workers are uncomfortable with the way their employers are tracking them, while others worry that AI will make their jobs obsolete. “Employers interested in investing in artificial intelligence systems must also invest in their employees, educating them about the role of AI and provide opportunities for feedback,” said Arthur Evans Jr., chief executive officer of APA. “The workplace is changing rapidly. Open and honest communication from employers can help relieve employees’ anxieties about the unknown and improve overall well-being, which is associated with higher organizational performance,” Evans said in an association news release. The APA poll surveyed more than 2,500 employed adults in the second half of April. Nearly 2 in 5 workers polled said they were worried that AI might one day make some or all of their job duties obsolete. About 64% of those who said they were worried about AI also said they felt tense or stressed during the workday, compared to 38% of those who were not worried about AI. Workers with a high school education or less were significantly more likely than those with a four-year college degree to be worried that their jobs would become obsolete — 44% to 34%. About 50% of… read on > read on >
Booming Sales of Legal Marijuana Linked to More Car Crashes
Emergency room visits for injuries related to driving under the influence of cannabis skyrocketed in Canada after the drug was legalized there, a new study reports. In October 2018, Canada became the second country to nationally legalize recreational or nonmedical cannabis for adult use. While known cannabis-involved emergency department (ED) visits for traffic injuries were still rare, they grew by 475% over 13 years, with a sharper rise in accidents after legalization, the researchers found. “Our findings highlight a concerning increase in cannabis-involvement in traffic-injury emergency visits over time, with even sharper spikes following the phases of legalization and commercialization,” said lead author Dr. Daniel Myran, a post-doctoral trainee at the Institute for Clinical Evaluative Sciences (ICES) and a family physician at the Ottawa Hospital. “Conversely, alcohol-involvement in traffic injury ED visits did not increase over the study period, which suggests that legalization of cannabis has played an important role in rising rates,” Myran said in an ICES news release. For the study, the researchers looked at cannabis-involved ED visits for traffic injuries between 2010 and 2021, looking for changes after the October 2018 commercialization of the legal cannabis market, which expanded products and retail stores. The investigators reviewed data from more than 947,000 ED visits for traffic injuries in the province of Ontario. Annual rates of cannabis-involved visits surged from 0.18 visits per 1,000… read on > read on >
AI Equals Human Radiologists at Interpreting Breast Cancer Scans
Another study is showing that artificial intelligence (AI) is as good as a specialist doctor in spotting breast cancer on a mammogram. But don’t expect computers to take over the job from humans, experts say. In a study that compared the mammography-reading skills of an AI tool with those of more than 500 medical professionals, researchers found that it was basically a tie. On average, both humans and AI caught about 90% of breast tumors, and correctly gave an all-clear to just over three-quarters of mammograms from women without cancer. That meant neither was perfect, and experts said it’s still unclear how AI will ultimately fit into breast cancer screening. Mammography has long been a routine experience for women. But mammography-reading may actually be the most challenging task in radiology, said Dr. Liane Philpotts, a professor of radiology at Yale School of Medicine in New Haven, Conn. That’s because a mammogram is an old-fashioned X-ray — although in the United States, Philpotts noted, better-performing digital 3D mammography is increasingly replacing the conventional kind. Detecting a tumor on standard mammograms means hunting for subtle patterns — something that has proved difficult even for the best AI, said Philpotts, who wrote an editorial published with the new findings in the September issue of Radiology. “In this study, we’re still talking about imperfect sensitivity,” Philpotts said, referring to… read on > read on >
Posting Online About Mental Health Could Harm Your Career: Study
It’s become more common for people to share mental health struggles on social media, but that decision could have a negative impact on future employment. Potential employers view job candidates differently if they talk about their personal mental health publicly, a new study finds. “People are often encouraged to discuss their mental health struggles on social media with the goal of reducing the stigma associated with mental health challenges,” said study co-author Lori Foster, a professor of psychology at North Carolina State University. “We think reducing stigma around mental health is extremely important, but our study suggests that mental health posts on platforms such as LinkedIn could have unforeseen consequences for people disclosing their mental health challenges,” Foster said in a university news release. The study showed these disclosures can influence the way employers view someone in professional contexts, said co-author Jenna McChesney, an assistant professor of psychology at Meredith College in Raleigh, N.C. She worked on the study while a grad student at NC State. “It’s important for people to take that into consideration when determining whether to share their mental health experiences online,” McChesney said in the release. The researchers enlisted 409 professionals with hiring experience to participate in the study, dividing them into four groups. One group was shown the LinkedIn page of a job candidate, with no mention of mental health… read on > read on >
Suicides Among U.S. Veterans Jumped 10-Fold in Decades After 9/11
Suicide has become an urgent issue among American military veterans, with rates increasing by more than 10 times in nearly two decades, a new study reveals. “Suicide rates for post-9/11 veterans have steadily increased over the last 15 years and at a much faster pace than the total U.S. population, and post-9/11 veterans with TBI [traumatic brain injuries] have a significantly higher suicide rate than veterans without TBI,” said lead researcher Jeffrey Howard, from the department of public health at the University of Texas at San Antonio. In fact, the suicide rate for those with a TBI was 56% higher than among veterans who didn’t suffer a TBI, the researchers found. Exposure to TBI, even a mild one, is associated with severe long-term health risks, including suicide, Howard noted. “It used to be believed that once initial symptoms of a mild TBI resolved, the patient was healed and there were no long-term health impacts, but as we are compiling longer-term follow-up data on these patients a different picture is emerging,” Howard said. These data suggest that closer and longer-term monitoring of patients with TBIs may be needed, he explained. “In addition to the clinical implications, the data point to the need for a more holistic approach to ensuring veterans’ health and well-being, which would integrate family and social support networks and other societal factors,” Howard… read on > read on >
Need Quick Help Learning CPR? Don’t Rely on Alexa, Siri
If you need quick directions on performing cardiopulmonary resuscitation (CPR) in an emergency, don’t rely on Alexa, Siri or another voice assistant. A new study finds the directions provided by these AI (artificial intelligence) helpers are inconsistent and lack relevance. “Our findings suggest that bystanders should call emergency services rather than relying on a voice assistant,” said co-author Dr. Adam Landman, chief information officer and senior vice president of digital operations at Mass General Brigham in Boston. “Voice assistants have potential to help provide CPR instructions, but need to have more standardized, evidence-based guidance built into their core functionalities,” Landman, an attending emergency physician, said in a hospital news release. Researchers presented eight verbal questions to four voice assistants: Amazon’s Alexa, Apple’s Siri, Google Assistant’s Nest Mini, and Microsoft’s Cortana. The study authors also typed the same questions into ChatGPT. The responses were evaluated by two board-certified emergency medicine physicians. Nearly half of the responses from the voice assistants were unrelated to CPR, the study found. This included information related to a movie called “CPR” and a link to Colorado Public Radio News. Only 28% of the replies suggested calling emergency services. Only 34% provided CPR instruction and just 12% gave verbal instructions. The most relevant information offered through AI was on ChatGPT. Using existing AI voice assistant tools may delay care and yield inappropriate… read on > read on >
Can You Rely on AI to Answer Questions About Cancer?
AI might not always be your most accurate source of health information, especially when it comes to cancer care, new research finds. Two new studies assessed the quality of responses offered by AI chatbots to a variety of questions about cancer care. One, published Aug. 24 in JAMA Oncology, zeroed in on the full-sentence conversational AI service known as ChatGPT, which launched to great fanfare last November. The upside: About two-thirds of cancer information offered by ChatGPT accurately matched current guidelines from the U.S. National Comprehensive Cancer Network. The downside: The rest did not. “Some recommendations were clearly completely incorrect,” said study author Dr. Danielle Bitterman, an assistant professor of radiation oncology at the Brigham and Women’s Hospital/Dana-Farber Cancer Institute and at Harvard Medical School in Boston. “For example, instances where curative treatment was recommended for an incurable diagnosis.” Other times, incorrect recommendations were more subtle — for instance, including some, but not all, parts of a treatment regimen, such as recommending surgery alone, when standard treatment also includes radiotherapy and/or chemotherapy, Bitterman said. That’s concerning, she said, given the degree to which “incorrect information was mixed in with correct information, which made it especially difficult to detect errors even for experts.” A second study in the same journal issue offered a much rosier assessment of AI accuracy. In this instance, investigators looked at answers… read on > read on >
Skipping Aspirin After Heart Attack Raises Odds for Recurrence
If you’ve had a heart attack, your doctor likely told you to take a low-dose aspirin daily to stave off a second heart attack or stroke, but most people don’t follow through with this advice over the long-term. Those folks who don’t take daily low-dose aspirin consistently are more likely to have another heart attack, stroke or die compared with their counterparts who consistently take aspirin, a new study shows. Aspirin keeps platelets from clumping together, which can help prevent or reduce the blood clots that can cause heart attacks and strokes. ”Most people should be on lifelong aspirin after a heart attack,” said Dr. Deepak Bhatt, director of Mount Sinai Heart and professor of cardiovascular medicine at the Icahn School of Medicine in New York City. “Long-term adherence to medication is a problem worldwide, including in the USA, and this is true even for inexpensive drugs such as aspirin, which can be life saving in heart attack patients,” said Bhatt, who had no role in the research. The study was led by Dr. Anna Meta Kristensen of Bispebjerg and Frederiksberg Hospital in Frederiksberg, Denmark. The researchers followed more than 40,100 people aged 40 or older who had a first-time heart attack from 2004 through 2017. The study team checked up on aspirin use two, four, six, and eight years after their heart attacks to… read on > read on >