ChatGPT's responses to people's healthcare-related queries are nearly indistinguishable from those provided by humans, a new study from NYU Tandon School of Engineering and Grossman School of Medicine reveals, suggesting the potential for chatbots to be effective allies to healthcare providers' communications with patients.
17 aug 2023--An NYU research team presented 392 people aged 18 and above with ten patient questions and responses, with half of the responses generated by a human healthcare provider and the other half by ChatGPT.
Participants were asked to identify the source of each response and rate their trust in the ChatGPT responses using a 5-point scale from completely untrustworthy to completely trustworthy.
The study found people have limited ability to distinguish between chatbot and human-generated responses. On average, participants correctly identified chatbot responses 65.5% of the time and provider responses 65.1% of the time, with ranges of 49.0% to 85.7% for different questions. Results remained consistent no matter the demographic categories of the respondents.
The study found participants mildly trust chatbots' responses overall (3.4 average score), with lower trust when the health-related complexity of the task in question was higher. Logistical questions (e.g. scheduling appointments, insurance questions) had the highest trust rating (3.94 average score), followed by preventative care (e.g. vaccines, cancer screenings, 3.52 average score). Diagnostic and treatment advice had the lowest trust ratings (scores 2.90 and 2.89, respectively).
According to the researchers, the study highlights the possibility that chatbots can assist in patient-provider communication particularly related to administrative tasks and common chronic disease management. Further research is needed, however, around chatbots' taking on more clinical roles. Providers should remain cautious and exercise critical judgment when curating chatbot-generated advice due to the limitations and potential biases of AI models.
The study, 'Putting ChatGPT's Medical Advice to the (Turing) Test: Survey Study,' is published in JMIR Medical Education.
More information: Oded Nov et al, Putting ChatGPT's Medical Advice to the (Turing) Test: Survey Study, JMIR Medical Education (2023). DOI: 10.2196/46939
Espresso coffee prevents Alzheimer's tau protein clumping in lab tests
Whether enjoyed on its own or mixed into a latte, Americano or even a martini, espresso provides an ultra-concentrated jolt of caffeine to coffee lovers. But it might do more than just wake you up. Research now published in the Journal of Agricultural and Food Chemistry shows that, in preliminary in vitro laboratory tests, espresso compounds can inhibit tau protein aggregation—a process that is believed to be involved in the onset of Alzheimer's disease.
17 aug 2023--Roughly half of all Americans drink coffee every day, and espresso is a popular way to consume it. To "pull" an espresso shot, hot water is forced through finely ground coffee beans, creating a concentrated extract. This is often used as a base for other drinks, including the trendy espresso martini.
Recent research has suggested that coffee could also have beneficial effects against certain neurodegenerative diseases, including Alzheimer's disease. Although the exact mechanisms that cause these conditions are still unclear, it's thought that a protein called tau plays a significant role.
In healthy people, tau proteins help stabilize structures in the brain, but when certain diseases develop, the proteins can clump together into fibrils. Some researchers propose that preventing this aggregation could alleviate symptoms. So, Mariapina D'Onofrio and colleagues wanted to see if compounds in espresso could prevent tau aggregation in vitro.
The researchers pulled espresso shots from store-bought beans, then characterized their chemical makeup using nuclear magnetic resonance spectroscopy. They chose caffeine and trigonelline, both alkaloids, the flavonoid genistein and theobromine, a compound also found in chocolate, to focus on in further experiments. These molecules, along with the complete espresso extract, were incubated alongside a shortened form of the tau protein for up to 40 hours.
As the concentration of espresso extract, caffeine or genistein increased, fibrils were shorter and didn't form larger sheets, with the complete extract showing the most dramatic results. Shortened fibrils were found to be non-toxic to cells, and they did not act as "seeds" for further aggregation.
In other experiments, the researchers observed that caffeine and the espresso extract could both bind pre-formed tau fibrils. Although much more research is needed, the team says that their preliminary in vitro findings could pave the way toward finding or designing other bioactive compounds against neurodegenerative diseases, including Alzheimer's.
More information: Espresso coffee mitigates the aggregation and condensation of Alzheimer's associated tau protein, Journal of Agricultural and Food Chemistry (2023). DOI: 10.1021/acs.jafc.3c01072
Older women at risk for Alzheimer's disease may benefit from yoga
Kundalini yoga, a form of yoga that focuses on breathing, meditation, and mental visualization, appeared beneficial for older women who had risk factors for Alzheimer's disease and concerns about episodes of memory decline, according to a UCLA Health study.
17 aug 2023--Researchers at UCLA Semel Institute for Neuroscience and Human Behavior, using a type of MRI that measures activity in regions and subregions of the brain, found that Kundaliniyoga, which combines movement and meditation and focuses on breathing, mantra recitation and mental visualization, increased connectivity in an area of the brain that can be impacted by stress and is associated withmemorydecline. Results have been published online in theJournal of Alzheimer's Disease.
Led by psychiatrist Dr. Helen Lavretsky, UCLA researchers studied the effects of yoga compared to the gold-standard approach of memory enhancement training (MET) on connectivity in subregions of the hippocampus, a critical area of the brain for learning and memory. MET is derived from techniques that use verbal and visual association and practical strategies to improve memory.
"Kundalini yoga training appears to better target stress-related hippocampal connectivity, whereas MET may better target sensory-integration subregions of the hippocampus, supporting better memory reliability," said Lavretsky, director of the Late-Life Mood, Stress, and Wellness Research Program.
"The key takeaway is that this study adds to the literature supporting the benefits of yoga for brain health, especially for women who have greater perceived stress and subjective memory impairment," she said. "This gentle form of yoga, which focuses more on breathing and mental engagement than on movement, like other forms of yoga, is ideal for older adults who may have some physical limitations."
The study included 22 participants who were part of a larger randomized controlled trial studying yoga's effects on Alzheimer's risk. Mean age among the 11 yoga participants was about 61; it was about 65 in the MET group. All had a self-reported decline in memory function during the previous year and one or more cardiovascular risk factors, which can also increase risk for Alzheimer's disease. These included plaque buildup in arteries, recent heart attack, diabetes, and treatment for high blood pressure or high cholesterol.
Both the yoga and MET groups had a 60-minute, in-person training session each week for 12 weeks. The programs also included daily homework or practice sessions. The Kundalini yoga (KY) training was supported with at-home practice of another brief meditative form of yoga, Kirtan Kriya (KK). These types of yoga engage a variety of senses simultaneously and have a chanting component that may improve respiratory, cardiovascular and autonomic nervous system functions, according to previous studies.
Lavretsky and her team have previously reported that Kundalini and Kirtan Kriya yoga had beneficial effects on depression, resilience and executive functioning in older adults with mild cognitive impairment. They also found that yoga had a more robust neuroprotective effect on right hippocampal volume—which may suggest improved memory function—than MET in older women with subjective memory decline and cardiovascular risk factors.
The new study used specialized functional MRI to establish resting-state connectivity of the hippocampus. This imaging, considered more sensitive to cognitive changes than hippocampal volumes, enabled the researchers to evaluate subregions of the hippocampus, comparing the effects of yoga versus memory training.
Based on their findings, the yoga "training may better target hippocampal subregion connectivity impacted by stress, which may aid in processing information, including facial information, into memory," the authors said. Additionally, "the observed greater increased connectivity between anterior and posterior hippocampal subregions with KY+KK training than with MET may suggest superior long-term neuroprotective benefits in terms of vulnerable hippocampal connections critical to episodic memory with KY+KK training."
The hippocampus must integrate information from various senses, and MET appears better than yoga in assisting in this function. "The various mnemonic strategies in MET, including verbal, visual, and spatial associative techniques, generally aim to enhance multimodal sensory integration into memory processes. Thus, MET may show superiority to KY+KK in terms of hippocampal sensory-integration important to memory," which could support better memory reliability, the study said.
While the small study suggests these forms of yoga may be of particular benefit to women who report experiencing stress and have additional risk factors for Alzheimer's disease, the authors say future, large-scale studies that have a placebo group or control arm will be needed to clarify the beneficial effects of both yoga and MET on hippocampal connectivity and memory.
More information: Lisa A. Kilpatrick et al, Impact of Yoga Versus Memory Enhancement Training on Hippocampal Connectivity in Older Women at Risk for Alzheimer's Disease, Journal of Alzheimer's Disease (2023). DOI: 10.3233/JAD-221159
Guidance on new treatments for early Alzheimer's disease issued
New therapies for early Alzheimer's disease, monoclonal antibodies that remove amyloid-β plaques in the brain, are bringing hope to people whose lives have been affected by the disease. To help neurologists discuss these therapies with patients and caregivers, the American Academy of Neurology has developed an Emerging Issues in Neurology article, published online on July 26, 2023, in Neurology.
17 aug 2023--Emerging Issues in Neurology articles are designed to provide timely guidance to neurologists and other clinicians, derived from expert consensus about new or emerging issues, that have immediate implications forpatient carebut for which a formal evidence base is still evolving.
"Neurologists care for millions of people with Alzheimer's disease and many people with early forms of dementia are eager to learn if these new therapies could help them," said American Academy of Neurology President Carlayne E. Jackson, MD, FAAN. "To help neurologists provide the highest quality care, experts with the American Academy of Neurology have summarized the available evidence on anti-amyloid monoclonal antibodies so that neurologists, patients and their caregivers can make informed decisions together about possible treatment with these therapies."
The Emerging Issues in Neurology article was written using available information on lecanemab, aducanumab and donanemab. It is important to note that this article is not a clinical practice guideline.
"Recent data on lecanemab and other monoclonal antibody infusions targeting amyloid-β protein make clear that new agents are highly likely to be part of the toolkit for neurologists caring for people with Alzheimer's disease," said article author Vijay K. Ramanan, MD, Ph.D., of the Mayo Clinic in Rochester, Minnesota. "While the formal evidence base is still evolving, this article was created with expert consensus until there is enough evidence on these therapies to inform evidence-based recommendations."
Lecanemab received traditional FDA approval on July 6, 2023. Aducanumab received accelerated approval from the FDA in June 2021 but has not yet received traditional approval. Aducanumab is currently available only to people participating in a clinical trial. Donanemab is not yet approved, but a decision on traditional FDA approval is expected later in 2023.
The article explains who is eligible to receive these therapies. Currently only people with early symptomatic forms of the disease, mild cognitive impairment or mild dementia due to Alzheimer's disease, may qualify to receive lecanemab. In addition, the article says people should be counseled about certain genetic risk factors and must not have a history of certain types of strokes. This is due to the risk of a serious side effect called amyloid-related imaging abnormalities, or ARIA, which is brain swelling and bleeding in the brain that can lead to death.
For this reason, people taking certain anticoagulant medications that are commonly prescribed to older adults may also not be eligible. There have been three deaths linked to lecanemab. The article notes that at least two of those people were given anticoagulants while on thetherapy.
While the goal of using these therapies is to remove amyloid-β plaques to slow cognitive decline, the article notes the therapies are not a cure for the disease. It also explains that the reduction in the rate of cognitive decline seen over 18 months in some studies may not be evident to the people receiving these therapies.
The article discusses the high cost of these therapies and notes that additional costs will come with diagnostic testing, administration and safety monitoring. The drugs are administered through regular infusions and monitoring requires multiple brain scans. Plus, there is a shortage of neurologists and medical professionals needed to provide this care and meet the anticipated demand.
The article expresses concern that study participants so far have primarily been white, while Black and Hispanic people have been underrepresented. It says steps must be taken to ensure that future studies include a diverse range of participants, especially since the incidence of dementia has been shown to be higher in Black and Hispanic populations compared with white populations.
"There is much optimism that anti-amyloid monoclonal antibodies may facilitate slowing of the disease process in some people with Alzheimer's disease," said Ramanan. "Additional research is needed to further determine who may be most likely to benefit from these therapies, as well as to find ways to improve outcomes for people using them and enable future advances in this new era of Alzheimer's disease care."
More information: Vijay K Ramanan et al, Antiamyloid Monoclonal Antibody Therapy for Alzheimer Disease: Emerging Issues in Neurology, Neurology (2023). DOI: 10.1212/WNL.0000000000207757
Nearly a half-million Americans might have tick-borne meat allergy syndrome
by Steven Reinberg
Tick bites can cause crippling infections like Lyme disease in humans, but new research suggests they can also trigger a serious meat allergy in far more Americans than thought.
17 aug 2023--Called alpha-gal syndrome, the condition may affect hundreds of thousands of Americans, U.S. health officials announced Thursday, but many doctors are not familiar with the condition, or how to diagnose or treat it.
According to one of two studies from researchers at the U.S. Centers for Disease Control and Prevention, both published in Morbidity and Mortality Weekly Report, there were more than 110,000 suspected cases of alpha-gal syndrome reported between 2010 and 2022. But because diagnosis requires a diagnostic test and a clinical exam, many people may not get tested. On that basis, the CDC researchers estimated that as many as 450,000 Americans might be affected.
In that study, researchers looked at lab results from 2017 to 2022 from a laboratory that—until August 2021—was the primary commercial lab offering such testing in the United States.
More than 300,000 samples were tested, and more than 30% showed AGS infection.
"Alpha-gal syndrome is an important emerging public health problem, with potentially severe health impacts that can last a lifetime for some patients," CDC researcher Dr. Ann Carpenter said in an agency news release. "It's critical for clinicians to be aware of AGS so they can properly evaluate, diagnose and manage their patients, and also educate them on tick bite prevention to protect patients from developing this allergic condition."
Unfortunately, a second study from the same CDC researchers found that many health care providers are not familiar with the potentially life-threatening allergic condition, the agency noted.
The survey of 1,500 family doctors, internists, pediatricians, nurse practitioners and physician assistants showed that nearly half (42%) had not heard of AGS, one-third said they were "not too confident" in their ability to diagnose or manage patients with the syndrome, and only 5% felt "very confident" in their ability.
Alpha-gal is a sugar found in meats like pork, beef, rabbit, lamb and venison, along with products made from gelatin, cow's milk, milk products and some pharmaceuticals. AGS is a serious allergic reaction some people get after eating food or products containing alpha-gal.
Evidence suggests that AGS is associated with the bite of a lone star tick, but other types of ticks have not been ruled out, the researchers said.
The Southern, Midwestern and mid-Atlantic regions have seen more people who test positive for AGS, the agency added.
"The burden of alpha-gal syndrome in the United States could be substantial, given the large percentage of cases suspected to be going undiagnosed due to non-specific and inconsistent symptoms, challenges seeking health care, and lack of clinician awareness," said CDC researcher Dr. Johanna Salzer, the senior author on both studies.
"It's important that people who think they may suffer from AGS see their health care provider or an allergist, provide a detailed history of symptoms, get a physical examination, and a blood test that looks for specific antibodies [proteins made by your immune system] to alpha-gal," she said in the release.
Symptoms of AGS can include hives or itchy rash, nausea or vomiting, heartburn or indigestion, diarrhea, cough, shortness of breath or difficulty breathing, drop in blood pressure, swelling of the lips, throat, tongue, or eyelids, dizziness or faintness, or severe stomach pain. Symptoms usually start two to six hours after eating meat or other exposure to products containing alpha-gal.
More information:Ann Carpenter et al, Health Care Provider Knowledge Regarding Alpha-gal Syndrome—United States, March–May 2022,MMWR. Morbidity and Mortality Weekly Report(2023).DOI: 10.15585/mmwr.mm7230a1
Julie M. Thompson et al, Geographic Distribution of Suspected Alpha-gal Syndrome Cases—United States, January 2017–December 2022, MMWR. Morbidity and Mortality Weekly Report (2023). DOI: 10.15585/mmwr.mm7230a2
Journal information: Morbidity and Mortality Weekly Report
Google & ChatGPT found to have mixed results in medical information queries
Screenshot of the evaluation web application. Each rater selects a response from the drop-down menus and then advances to the next question. Credit: Journal of Medical Internet Research (2023). DOI: 10.2196/48966
When you need accurate information about a serious illness, should you go to Google or ChatGPT?
17 aug 2023--Aninterdisciplinary studyled by University of California, Riverside,computer scientistsfound that both internet information gathering services have strengths and weaknesses for people seeking information about Alzheimer's disease and other forms of dementia. The team included clinical scientists from the University of Alabama and Florida International University.
Google provides the most current information, but query results are skewed by service and product providers seeking customers, the researchers found. ChatGPT, meanwhile, provides more objective information, but it can be outdated and lacks the sources of its information in its narrative responses.
"If you pick the best features of both, you can build a better system, and I think that this is what will happen in the next couple of years," said Vagelis Hristidis, a professor of computer science and engineering in UCR's Bourns College of Engineering.
In their study, Hristidis and his co-authors submitted 60 queries to both Google and ChatGPT that would be typical submissions from people living with dementia and their families.
The researchers focused on dementia because more than 6 million Americans are impacted by Alzheimer's disease or a related condition, said study co-author Nicole Ruggiano, a professor of social work at the University of Alabama.
"Research also shows that caregivers of people living with dementia are among the most engaged stakeholders in pursuing health information, since they often are tasked with making decisions for their loved one's care," Ruggiano said.
Half of the queries submitted by the researchers sought information about the disease processes, while the other half sought information on services that could assist patients and their families.
The results were mixed.
"Google has more up-to-date information, and covers everything," Hristidis said. "Whereas ChatGPT is trained every few months. So, it is behind. Let's say there's some new medicine that just came out last week, you will not find it on ChatGPT."
While dated, ChatGPT provided more reliable andaccurate informationthan Google. This is because the ChatGPT creators at OpenAI choose the most reliable websites when they train ChatGPT through computationally intensive machine learning. Yet, users are left in dark about specific sources of information because the resulting narratives are void of references.
Google, however, has a reliability problem because it essentially "covers everything from the reliable sources to advertisements," Hristidis said.
In fact, advertisers pay Google for their website links to appear at the top of search result pages. So, users often first see links to websites of for-profit companies trying to sell them care-related services and products. Finding reliable information from Google searches thus requires a level of user skill and experience, Hristidis said.
Co-author Ellen Brown, an associate professor of nursing at the Florida International University, pointed out that families need timely information about Alzheimer's. .
"Although there is no cure for the disease, many clinical trials are underway and recently a promising treatment for early stage Alzheimer's disease was approved by the FDA," Brown said. "Therefore, up-to-date information is important for families looking to learn about recent discoveries and available treatments."
The authors of the study write that "the addition of both the source and the date of health-related information and availability in other languages may increase the value of these platforms for both non-medical and medical professionals." It was published in the Journal of Medical Internet Research under the title "ChatGPT vs. Google for Queries Related to Dementia and Other Cognitive Decline: Comparison of Results."
Google and ChatGPT both scored low for readability scores, which makes it difficult for people with lower levels of education and low health literacy skills.
"My prediction is that the readability is the easier thing to improve because there are already some tools, some AI methods, that can read and paraphrase text," Hristidis said. "In terms of improving reliability, accuracy, and so on, that's much harder. Don't forget that it took scientists many decades of AI research to build ChatGPT. It is going to be slow improvements from where we are now."
More information: Vagelis Hristidis et al, ChatGPT vs Google for Queries Related to Dementia and Other Cognitive Decline: Comparison of Results, Journal of Medical Internet Research (2023). DOI: 10.2196/48966
Journal information: Journal of Medical Internet Research
Constipation associated with increased risk of hypertension, cardiovascular events in elderly Australian patients
Summary of the findings. This study found that constipation was positively associated with hypertension and cardiovascular events. CI, confidence interval; OR, odds ratio. This figure was partly generated using Servier Medical Art, provided by Servier, licensed under a Creative Commons Attribution 3.0 unported license. Credit: Scientific Reports (2023). DOI: 10.1038/s41598-023-38068-y
A relatively common health problem—constipation—has been shown to be a risk factor for hypertension and cardiovascular events such as stroke in people over 60.
17 aug 2023--The extensive La Trobe University study of over half a millionhospital admissionsin Victoria suggests that interventions to addressconstipationmay reduce these risks inelderly patients.
The study, led by Professors Grant Drummond and Chris Sobey and published in Scientific Reports, looked at 541,172 hospitalized patients over 60 years of age.
For each constipation admission, one exact age-matched non-constipated admission was randomly selected from all hospitalizations within two weeks to form the control arm of the study.
The researchers found that patients with constipation had almost double the risk for hypertension, and were also more likely to suffer from major cardiovascular events such as heart attack and stroke.
The number of people with cardiovascular diseases (CVD) has nearly doubled over the past 30 years, and the number of deaths from CVD has increased from 12.1 million to 18.6 million during this time.
According to Professor Drummond, despite efforts to modify traditional risk factors for CVD with lifestyle and drug interventions, cardiovascular events are still responsible for 32% of global deaths, 85% of which are due to heart attacks or stroke.
"Therefore, identifying non-traditional CVD risk factors and developing strategies to address them is critical to further reduce CVD-associated morbidity and mortality," Professor Drummond said.
While it is unclear whether constipation is a direct the cause of hypertension in elderly patients, according to Professor Sobey,
"Such a relationship is plausible because in constipation there is increased water absorption from the gut, microbiota changes, and inflammation, all of which could lead to hypertension," Professor Sobey said.
The study found that:
constipation in patients was associated with a 96% increased risk of hypertension and also a an increased risk of myocardial infarction, stroke and all cardiovascular events, compared to patients with no constipation
patients with both constipation and hypertension had a more than 500% higher risk of cardiovascular events than patients with neither condition.
these relationships were similar in males and females
15.2% of people over 60 in the study suffered from constipation
More information: Courtney P. Judkins et al, Association of constipation with increased risk of hypertension and cardiovascular events in elderly Australian patients, Scientific Reports (2023). DOI: 10.1038/s41598-023-38068-y
World's largest study shows the more you walk, the lower your risk of death, even if you walk fewer than 5,000 steps
The number of steps you should walk every day to start seeing benefits to your health is lower than previously thought, according to the largest analysis to investigate this.
17 aug 2023--The study, published in theEuropean Journal of Preventive Cardiologytoday, found that walking at least 3967 steps a day started to reduce the risk of dying from any cause, and 2337 steps a day reduced the risk of dying from diseases of the heart and blood vessels (cardiovascular disease).
However, the new analysis of 226,889 people from 17 different studies around the world has shown that the more you walk, the greater the health benefits. The risk of dying from any cause or from cardiovascular disease decreases significantly with every 500 to 1000 extra steps you walk. An increase of 1000 steps a day was associated with a 15% reduction in the risk of dying from any cause, and an increase of 500 steps a day was associated with a 7% reduction in dying from cardiovascular disease.
The researchers, led by Maciej Banach, Professor of Cardiology at the Medical University of Lodz, Poland, and Adjunct Professor at the Ciccarone Center for the Prevention of Cardiovascular Disease, Johns Hopkins University School of Medicine, found that even if people walked as many as 20,000 steps a day, the health benefits continued to increase. They have not found an upper limit yet.
"Our study confirms that the more you walk, the better," says Prof. Banach. "We found that this applied to both men and women, irrespective of age, and irrespective of whether you live in a temperate, sub-tropical or sub-polar region of the world, or a region with a mixture of climates. In addition, our analysis indicates that as little as 4,000 steps a day are needed to significantly reduce deaths from any cause, and even fewer to reduce deaths from cardiovascular disease."
There is strong evidence that a sedentary lifestyle may contribute to an increase in cardiovascular disease and a shorter life. Studies have shown that insufficient physical activity affects more than a quarter of the world's population. More women than men (32% versus 23%), and people in higher income countries compared to low-income countries (37% versus 16%) do not undertake a sufficient amount of physical activity.
According to World Health Organization data, insufficient physical activity is the fourth most frequent cause of death in the world, with 3.2 million deaths a year related to physical inactivity. The COVID-19 pandemic also resulted in a reduction in physical activity, and activity levels have not recovered two years on from it.
Dr. Ibadete Bytyçi from the University Clinical Centre of Kosovo, Pristina, Kosovo, senior author of the paper, says, "Until now, it's not been clear what is the optimal number of steps, both in terms of the cut-off points over which we can start to see health benefits, and the upper limit, if any, and the role this plays in people's health. However, I should emphasize that there were limited data available on step counts up to 20,000 a day, and so these results need to be confirmed in larger groups of people."
This meta-analysis is the first not only to assess the effect of walking up to 20,000 steps a day, but also to look at whether there are any differences depending on age, sex or where in the world people live.
The studies analyzed by the researchers followed up participants for a median (average) of seven years. The mean (average) age was 64, and 49% of participants were female.
In people aged 60 years or older, the size of the reduction in risk of death was smaller than that seen in people aged younger than 60 years. In the older adults, there was a 42% reduction in risk seen in those who walked between 6,000 and 10,000 steps a day, while there was a 49% reduction in risk in younger adults who walked between 7,000 and 13,000 steps a day.
Prof. Banach says, "In a world where we have more and more advanced drugs to target specific conditions such as cardiovascular disease, I believe we should always emphasize that lifestyle changes, including diet and exercise, which was a main hero of our analysis, might be at least as, or even more effective in reducing cardiovascular risk and prolonging lives."
"We still need good studies to investigate whether these benefits may exist for intensive types of exertion, such as marathon running and iron man challenges, and in different populations of different ages, and with different associated health problems. However, it seems that, as with pharmacological treatments, we should always think about personalizing lifestyle changes."
Strengths of the meta-analysis include its size and that it was not restricted to looking at studies limited to a maximum of 16,000 steps a day. Limitations include that it was an observational study and so cannot prove that increased step counts cause the reduction in the risk of death, only that it is associated with it. The impact of step counts was not tested on people with different diseases; all the participants were generally healthy when they entered the studies analyzed.
The researchers were not able to account for differences in race and socioeconomic status, and the methods for counting steps were not identical in all the studies included in this meta-analysis.
More information: Maciej Banach et al, The Association Between Daily Step Count and All-Cause and Cardiovascular Mortality: A Meta-Analysis, European Journal of Preventive Cardiology (2023). DOI: 10.1093/eurjpc/zwad229
Medical writing: Caution warranted if using ChatGPT
Distribution of the references provided by ChatGPT (n=59). Credit: Mayo Clinic Proceedings: Digital Health (2023). DOI: 10.1016/j.mcpdig.2023.05.004
When it comes to health care, it's best to ask a professional. This oft-repeated adage also applies to scientists who might be tempted to use the ChatGPT artificial intelligence model for medical writing.
17 aug 2023--Researchers from CHU Sainte-Justine and the Montreal Children's Hospital recently posed 20 medical questions to ChatGPT. The chatbot provided answers of limited quality, including factual errors and fabricated references, show the results of the study published inMayo Clinic Proceedings: Digital Health.
"These results are alarming, given that trust is a pillar of scientific communication. ChatGPT users should pay particular attention to the references provided before integrating them into medical manuscripts," says Dr. Jocelyn Gravel, lead author of the study and emergency physician at CHU Sainte-Justine.
Striking findings
The researchers drew their questions from existing studies and asked ChatGPT to support its answers with references. They then asked the authors of the articles from which the questions were taken to rate the software's answers on a scale from 0 to 100%.
Out of 20 authors, 17 agreed to review the answers of ChatGPT. They judged them to be of questionable quality (median score of 60%). They also found major (five) and minor (seven) factual errors. For example, the software suggested administering an anti-inflammatory drug by injection, when it should be swallowed. ChatGPT also overestimated the global burden of mortality associated with Shigella infections by a factor of ten.
Of the references provided, 69% were fabricated, yet looked real. Most of the false citations (95%) used the names of authors who had already published articles on a related subject, or came from recognized organizations such as the Centers for Disease Control and Prevention or the Food and Drug Administration. The references all bore a title related to the subject of the question and used the names of known journals or websites.
Even some of the real references contained errors (eight out of 18).
ChatGPT explains
When asked about the accuracy of the references provided, ChatGPT gave varying answers. In one case, it claimed, "References are available in Pubmed," and provided a web link. This link referred to other publications unrelated to the question. At another point, the software replied, "I strive to provide the most accurate and up-to-date information available to me, but errors or inaccuracies can occur."
"The importance of proper referencing in science is undeniable. The quality and breadth of the references provided in authentic studies demonstrate that the researchers have performed a complete literature review and are knowledgeable about the topic. This process enables the integration of findings in the context of previous work, a fundamental aspect of medical research advancement. Failing to provide references is one thing but creating fake references would be considered fraudulent for researchers," says Dr. Esli Osmanlliu, emergency physician at the Montreal Children's Hospital and scientist with the Child Health and Human Development Program at the Research Institute of the McGill University Health Centre.
"Researchers using ChatGPT may be misled by false information because clear, seemingly coherent and stylistically appealing references can conceal poor content quality," adds Dr. Osmanlliu.
This is the first study to assess the quality and accuracy of references provided by ChatGPT, the researchers point out.
More information: Jocelyn Gravel et al, Learning to Fake It: Limited Responses and Fabricated References Provided by ChatGPT for Medical Questions, Mayo Clinic Proceedings: Digital Health (2023). DOI: 10.1016/j.mcpdig.2023.05.004
