Background: Adolescent anxiety is a growing public health concern associated with significant social and emotional impairment. Mindfulness-based interventions (MBIs) have shown promise in reducing anxiety and improving well-being; however, engagement remains challenging. Virtual reality (VR)–based delivery may enhance immersion and attention, potentially addressing barriers of traditional mindfulness formats. Evidence on VR-based mindfulness interventions for adolescents, particularly in Hong Kong, remains limited. Objective: This study aimed to evaluate the feasibility and acceptability of a VR-MBI delivered via a CAVE, an enclosed VR environment with three projected walls displaying immersive natural scenes and ambient sounds, for adolescents with mild-to-moderate anxiety symptoms in Hong Kong. Secondary aims were to explore preliminary effects on psychological outcomes and physiological stress regulation and to identify facilitators and barriers to engagement. Methods: A mixed methods, single-group pre-post study was conducted with adolescents experiencing mild-to-moderate anxiety symptoms, recruited from secondary schools and youth service organizations in Hong Kong. Participants completed an 8-week group-based VR-MBI. Feasibility and acceptability were assessed using recruitment, attendance, retention, homework practice frequency, dropouts, and adverse events. Psychological outcomes were measured using the Depression Anxiety Stress Scale–21 and the Mindful Attention Awareness Scale. Heart rate variability indices, including the standard deviation of normal-to-normal intervals and root-mean-square of successive differences, were collected at baseline and postintervention using a wearable device. Focus group interviews explored participants’ experiences. Paired-sample tests and Wilcoxon signed rank tests examined pre-post changes, and qualitative data were analyzed using thematic analysis, with findings integrated through triangulation. Results: A total of 42 participants (mean age 14.88, SD 1.90 years; 20/42, 47.6% female; 22/42, 52.4% male) enrolled and completed both assessments. Attendance was high, with 73.8% (31/42) of participants attending at least 80% (8/10) sessions, and participants engaged in regular homework practice. No dropouts or adverse events were reported. No significant pre-post changes were observed in self-reported distress, anxiety, depression, stress, or trait mindfulness (all >.05). However, significant improvements were observed in both heart rate variability indices, standard deviation of normal-to-normal intervals (mean difference 17.6 ms, 95% CI −33.88 to −1.32; =.04; Cohen =0.38) and root-mean-square of successive differences (mean difference 20.20 ms, 95% CI −38.76 to −1.65; =.03; Cohen =0.39), which may suggest preliminary enhancements in physiological stress regulation. Qualitative findings suggested perceived benefits in emotional regulation, stress reduction, focus, and sleep, with the immersive environment and group-based format identified as key facilitators. Conclusions: The CAVE-based VR-MBI was feasible and acceptable for adolescents with mild-to-moderate anxiety symptoms in Hong Kong. Despite no significant changes in self-reported outcomes, physiological improvements and positive qualitative feedback suggest early benefits not captured by self-report measures. These findings support further investigation of using controlled designs and longer follow-up periods.
Background: University students experience elevated psychological distress, with limited access to mental health services. While cognitive behavioral therapy (CBT) demonstrates efficacy for anxiety and depression, treatment gaps persist due to access barriers and insufficient between-session support. Large language model (LLM) chatbots could improve and scale CBT delivery. However, the scientific evaluation of chatbot-enhanced protocols is just emerging. Objective: This pilot study aimed to assess the feasibility, acceptability, and preliminary efficacy of an LLM-based ChatBot as an adjunct to group Unified Protocol (UP) therapy for between-session support in university students with subclinical anxiety and depression symptoms. Methods: A single-arm feasibility trial recruited university students aged 18 years and older with moderate subclinical symptoms (Social Phobia Inventory: 21‐40, Patient Health Questionnaire-9: 5‐14, or Generalized Anxiety Disorder-7: 5‐14), excluding those with current psychiatric disorders, suicidal ideation, or psychotropic medication use. The intervention comprised 4 weekly group UP counseling sessions complemented by an adjunctive Claude 3.7-Sonnet LLM ChatBot programmed with UP-based therapeutic prompts for between-session support rather than a stand-alone therapeutic agent. Primary feasibility outcomes included treatment adherence, chatbot engagement metrics, and system usability (System Usability Scale). Secondary outcomes assessed changes in generalized anxiety (Generalized Anxiety Disorder-7 Scale), social anxiety (Social Phobia Inventory), depression (Patient Health Questionnaire-9), and well-being (Short Warwick-Edinburgh Mental Wellbeing Scale) using paired tests. Qualitative feedback was collected through focus group interviews and analyzed using thematic analysis. Results: Of 72 screened participants, 37 met eligibility criteria and 19 initiated treatment (mean age 22.06, SD 1.78 years; 70.6% female). Retention was high with 17 completers (10.5% dropout rate). Among completers, 94.1% (16/17) attended ≥3 group sessions. The engagement with the CBT ChatBot was substantial: participants were active on a median of 23 days during the 34-day study period and exchanged a median of 15 messages in total. System usability was rated as excellent (mean 84.94, SD 10.98 out of 100). Pre-to-post comparisons revealed significant improvements in generalized anxiety (mean change −3.00, SD 3.46; =3.01, =.004; Cohen =0.71) and mental well-being (mean change +2.29, SD 3.65; =−2.17, =.02; Cohen =0.69). Social anxiety and depression showed nonsignificant trends toward improvement. Qualitative feedback highlighted the CBT ChatBot’s accessibility and nonjudgmental support while noting limitations in personalization. No adverse events or inappropriate chatbot interactions occurred. Conclusions: Augmenting a group UP therapy with an LLM ChatBot demonstrated high feasibility, acceptability, and preliminary efficacy signals for university students with subclinical symptoms. The hybrid intervention package achieved strong retention and engagement while maintaining safety. These findings support progression to a randomized controlled trial to definitively evaluate this technology-enhanced approach for expanding access to evidence-based mental health interventions.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/ceb992b5356218740d97471bbda48397" />
<strong>Background:</strong> Delirium superimposed on dementia is associated with poor outcomes yet remains underdetected in home settings. Current detection relies on face-to-face clinical assessment (eg, the Confusion Assessment Method criteria), which is rarely applied outside hospitals. <strong>Objective:</strong> This proof-of-concept study developed a theory-driven framework for detecting delirium-consistent anomalous patterns in home-dwelling people with dementia, using passive smart home sensor data. <strong>Methods:</strong> The Technology Integrated Health Management dataset, an open access resource comprising a clinically derived cohort of older adults (aged 50 years) with a confirmed diagnosis of dementia or mild cognitive impairment, was used. The analysis included 13 patients who had at least 50% valid data for at least one 10-day analysis window, with data collected between April 1, 2019, and June 30, 2019. Individualized anomaly detection algorithms, including Isolation Forest and Long Short-Term Memory models, were applied to identify delirium-related anomalies within each participant. Predictor features consisted of theory-driven digital markers approximating key Confusion Assessment Method criteria, including agitation, disrupted sleep-wake cycles, and disorientation (indexed by activity entropy), along with clinically relevant indicators, such as physiological instability (early warning scores) and urinary tract infections. <strong>Results:</strong> Using matched thresholds, the Isolation Forest identified 77 anomalies (anomaly rate: 15.65%), and the Long Short-Term Memory model identified 78 anomalies (anomaly rate: 15.85%), with anomalies typically occurring in short temporal clusters; agreement between methods ranged from 0% to 40% across individuals. Feature importance analyses indicated that activity entropy, sleep quality, and early warning scores were the most influential features, with stronger interfeature correlations observed during anomaly periods than during nonanomaly periods. <strong>Conclusions:</strong> This study demonstrates the technical feasibility of detecting delirium-related anomalies through passive smart home monitoring. While lacking ground truth validation, the approach shows promise for early intervention in community settings. Future validation studies with clinically confirmed delirium labels are essential. <strong>Trial Registration:</strong>
<strong>Background:</strong> Fluid assessment in geriatric inpatients is challenging, as clinical signs are often unreliable. Inferior vena cava (IVC) ultrasound provides a rapid, noninvasive estimation of intravascular volume. Teleguided point-of-care ultrasound (POCUS) allows examiners without prior ultrasound experience to perform scans under real-time supervision. <strong>Objective:</strong> This study aimed to evaluate the feasibility, accuracy, efficiency, and user satisfaction of remote-guided IVC ultrasound performed by medical students and nurses without prior ultrasound experience in a geriatric inpatient setting. <strong>Methods:</strong> This prospective feasibility study was conducted between February and March 2025 in a geriatric inpatient ward at a German tertiary care hospital. Thirty hospitalized geriatric patients were recruited using a pragmatic convenience sampling approach on predefined study days. Each patient underwent 2 IVC ultrasound examinations (n=60) using a handheld device with TeleGuidance; one was performed by a medical student and one by a nurse. All scans were remotely supervised by an ultrasound-experienced cardiologist, who subsequently performed a third, independent IVC scan on each patient, serving as the reference standard. Examiners were 2 final-year medical students and 2 nurses, all without ultrasound experience, each performing 15 scans. Primary outcomes were technical feasibility (successful teleguidance connection), accuracy of IVC diameter measurement (≥80% within +2 mm to –2 mm), and examination duration (≤10 minutes). The secondary outcome was user satisfaction (≥75 on a 0-100 numeric rating scale). <strong>Results:</strong> Connectivity and remote supervision were consistently stable, enabling completion of all scans (feasibility 100%). IVC visualization was successful in 90% (27/30) of cases. Accuracy was achieved in 80% (48/60; 95% CI 67-88) of scans. Mean duration was 3.3 (SD 2.0) minutes. Mean user satisfaction was 89%, with all ratings ≥85%. <strong>Conclusions:</strong> Telemedicine-guided IVC ultrasound was feasible and well accepted in this geriatric inpatient setting. Nonexpert examiners were able to obtain clinically usable measurements under remote supervision within a few minutes after minimal training. These findings suggest that teleguided POCUS is a promising approach to support task sharing in geriatric care. Further studies are needed to confirm these results and to evaluate integration into clinical practice. <strong>Trial Registration:</strong> German Clinical Trials Register DRKS00035821; https://www.drks.de/search/de/trial/DRKS00035821/details
Background: Digital meditation-based interventions (MBIs) reach vast global audiences with millions of active users, yet concerns persist about the frequency and nature of adverse experiences (ie, AExs) occurring during meditation training. Some researchers have argued that AExs are substantially underdetected and reflect iatrogenic harm caused by meditation (ie, adverse effects [AEfs]). Others contend that these experiences largely reflect common stressors that would be experienced without meditation. These competing perspectives underscore the need for further research, particularly in the context of digital MBIs, the most widely used form of meditation training. Objective: This study examined the prevalence, predictors, and subjective evaluations of AExs during a digital MBI and tested whether reported experiences may be caused by meditation practice via comparisons between meditation-exposed and nonexposed participants. Methods: Data were drawn from 2 trials of the Healthy Minds Program. Exploratory study 1 (n=315) consisted of a sample of distressed US undergraduate students to estimate the prevalence of AExs and identify baseline predictors. Preregistered confirmatory study 2 (n=594) sampled distressed US adults from all 50 states to replicate findings from study 1 and to examine participants’ subjective evaluations of AExs. Study 2 additionally compared AEx rates between participants who did and did not complete guided meditations to assess whether AExs could be caused by meditation exposure. Study 3 (n=87) used qualitative methods to analyze study 1 participants’ responses to an open-ended question regarding their strategies for coping with AExs. Results: In studies 1 and 2, 27.9% (88/315) and 10.1% (40/396) of participants, respectively, reported at least one AEx during the study period, with 6.7% (21/315) and 3% (12/396) reporting functional impairment, largely aligning with previous research. Critically, in study 2, rates of AExs did not significantly differ between participants who did and did not complete guided meditations, suggesting that these experiences were not caused by meditation practice. Higher baseline depression, anxiety, loneliness, experiential avoidance, and perceived barriers to meditation predicted more frequent AExs. In studies 1 and 2, 89.8% (79/88) and 90% (36/40) of participants who reported AExs, respectively, indicated that they were glad to have learned to meditate. Qualitative analyses showed that participants used diverse coping strategies, often using skills learned through the Healthy Minds Program. Conclusions: AExs were relatively common but occurred at comparable rates among participants who did and did not meditate, challenging claims that such experiences were caused by meditation practice in distressed individuals. Although a small subset of participants reported some degree of functional impairment, most evaluated their AExs as tolerable and described their overall MBI experience as positive. Together, these findings highlight the importance of distinguishing AExs that likely reflect epiphenomena of preexisting distress or symptoms from iatrogenic harm attributable to MBIs. Trial Registration: Study 1: ClinicalTrials.gov NCT04741529; https://clinicaltrials.gov/study/NCT04741529; Study 2: ClinicalTrials.gov NCT06282523; https://clinicaltrials.gov/study/NCT06282523
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/09106e6bad7f3f15798304e0c00626ec" />
WASHINGTON — The Trump administration on Friday proposed to change a policy that is designed to prevent drugmakers from avoiding Medicare price negotiation by adding active ingredients to drugs.
The policy is part of an annual proposed rule that establishes the process that the Centers for Medicare and Medicaid Services uses to choose the next 20 drugs and biologics for price negotiation. Those drugs will be announced by Feb. 1, 2027, and their negotiated prices will take effect in 2029. The administration also considered a similar policy last year but put off a decision to study it further.
Medicare must wait seven to 11 years after a product is approved by the Food and Drug Administration before it can negotiate its price, depending on the type of medicine. Biologics that are typically administered in doctor offices get more time than drugs taken orally.
Researchers have uncovered evidence that age-related genetic mutations commonly associated with cancer and blood disorders may also contribute to the chronic brain inflammation characteristic of Alzheimer’s disease.
In a study published in Cell, investigators from the Icahn School of Medicine at Mount Sinai and Boston Children’s Hospital identified somatic mutations in brain immune cells from patients with Alzheimer’s disease. The findings suggest that genetic alterations acquired during aging may reshape the behavior of these cells, driving inflammatory processes linked to neurodegeneration.
The work introduces a potential new biological mechanism connecting aging, immune dysfunction, and Alzheimer’s disease, a condition that affects millions of people worldwide and remains one of the leading causes of dementia.
Linking aging, immunity, and neurodegeneration
Inflammation has long been recognized as a central feature of Alzheimer’s disease. Activated immune cells are commonly observed surrounding amyloid plaques and other pathological hallmarks of the disease. However, the factors that initiate and sustain these harmful immune responses have remained incompletely understood.
The new study focused on microglia, the brain’s resident immune cells. These specialized cells play critical roles in maintaining neural health by clearing cellular debris, regulating synaptic connections, responding to injury, and coordinating immune activity within the central nervous system.
Previous research has shown that microglia can adopt disease-associated states during Alzheimer’s progression, but the molecular triggers responsible for these changes have been unclear.
“Our data suggest that some immune cells in Alzheimer’s disease undergo genetic changes over time that alter their behavior and potentially contribute to chronic inflammation in the brain,” said senior author Samuele Marro, PhD, associate professor in the Nash Family Department of Neuroscience and The Friedman Brain Institute at Mount Sinai.
“These findings provide a new framework for understanding how aging, immune dysfunction, and neurodegeneration may intersect in Alzheimer’s disease.”
Large-scale analysis of Alzheimer’s brains
To investigate whether acquired mutations might contribute to disease pathology, researchers analyzed 311 postmortem brain samples from individuals with Alzheimer’s disease and age-matched controls.
Using ultra-deep sequencing, the team screened 149 genes frequently associated with cancer and clonal hematopoiesis, an age-related condition in which blood stem cells acquire mutations that allow certain cell populations to expand disproportionately over time.
The analysis revealed significantly higher numbers of somatic mutations in Alzheimer’s disease brains compared with controls.
Several of the most frequently mutated genes, including TET2, DNMT3A, and ASXL1, are well-known drivers of clonal hematopoiesis and have previously been implicated in age-related blood disorders and cancer development.
Many of these mutations were highly enriched in microglia-like immune cells while being largely absent from neurons, suggesting that immune cells may be a primary target of these age-related genetic alterations.
Evidence for blood-derived immune cell involvement
The researchers also examined matched blood samples from some patients and found that many of the same mutations detected in the brain were present in circulating blood cells.
This observation suggests a possible route by which mutated immune cells originating from the blood may enter the brain and adopt microglia-like functions.
Growing evidence has indicated that peripheral immune cells can contribute to neuroinflammation under certain conditions. The new findings raise the possibility that age-related expansion of mutated blood cell clones could influence inflammatory processes within the brain.
Such a mechanism would provide a direct biological link between clonal hematopoiesis—an increasingly recognized consequence of aging—and neurodegenerative disease.
Functional effects of Alzheimer’s-associated mutations
To determine whether these mutations actively alter immune cell behavior, the researchers combined single-cell analyses with stem cell-based experimental models.
Using CRISPR gene editing, they engineered induced pluripotent stem cell-derived microglia-like cells carrying mutations identified in Alzheimer’s disease samples.
The resulting cells displayed profound changes in gene expression, adopting inflammatory programs and disease-associated microglial states that have previously been linked to neurodegeneration.
“Our study provides functional evidence that mutations commonly associated with aging blood cells and cancer biology can directly alter the behavior of brain immune cells,” said co-corresponding author Eirini Papapetrou, MD, PhD, professor of oncological sciences at Mount Sinai and director of the Center for Advancement of Blood Cancer Therapies.
“These mutated cells showed inflammatory signatures strongly associated with neurodegeneration.”
The findings suggest that these mutations are not merely bystanders but may actively influence cellular pathways involved in disease progression.
A potential new contributor to Alzheimer’s pathology
The study expands the growing view of Alzheimer’s disease as a disorder involving complex interactions between the immune system and the nervous system.
Historically, Alzheimer’s research has focused heavily on amyloid-beta plaques and tau tangles. More recently, attention has shifted toward the role of neuroinflammation and immune dysfunction as key drivers of disease progression.
The identification of somatic mutations in microglia-like cells adds another layer to this emerging picture. Rather than being solely inherited or environmentally driven, some aspects of Alzheimer’s pathology may arise from genetic alterations that accumulate naturally with age.
Because clonal hematopoiesis becomes increasingly common in older adults, the findings may have implications beyond Alzheimer’s disease and could influence understanding of other neurodegenerative disorders characterized by chronic inflammation.
Implications for future therapies
Although the study does not establish that these mutations directly cause Alzheimer’s disease, it identifies a plausible mechanism through which age-related genetic changes could exacerbate neurodegeneration.
Future studies will be needed to determine how early these mutations emerge, whether they predict disease risk, and whether interventions targeting mutated immune cell populations could slow disease progression.
“This work highlights a potentially important connection between aging blood biology and neurodegenerative disease,” said Marro.
“If confirmed in future studies, these findings could open new avenues for therapies that target harmful inflammatory immune cell populations in the brain.”
The researchers are now planning follow-up studies in animal models to further investigate the role of mutated immune cells in Alzheimer’s disease and evaluate whether reducing their inflammatory activity can modify the course of neurodegeneration.
Scientific publishing is facing a growing challenge from fabricated research produced by industrial-scale paper mills. But researchers and publishers are fighting back through technology and collaboration to protect the integrity of the scientific record.
Scientific publishing is based on the trust that the data are real and that peer review ensures quality. But that trust is being eroded by commercial enterprises known as paper mills—coordinated commercial operations that sell authorship slots in fraudulent or manipulated manuscripts, then submit those manuscripts to journals.
Unlike traditional misconduct, these are not lone researchers cutting corners but businesses producing research at scale, often tailored to meet the demands of specific fields, journals, and career incentives.
A recent analysis of almost 19,000 online adverts for paper mills revealed authorship slots being sold for between $36 to $5,600 depending on the position of the slot, highlighting how commercialized the market is. The average for a first author position was $1,030 and, although the study did not examine which adverts resulted in published papers, another investigation traced approximately 1,000 authorship adverts to more than 400 published papers.
The magnitude of the problem is difficult to estimate. A 2022 report by the Committee on Publication Ethics (COPE) and the International Association of Scientific, Technical & Medical Publishers (STM) found that the percentage of suspect papers submitted to journals was around two percent overall but increased sharply to as high as 46% in journals targeted by paper mills.
The pressure to publish
Adrian Barnett, PhD Professor Queensland University of Technology
For honest researchers, it might be difficult to understand why paper mills even exist. At the heart of the issue is what Adrian Barnett, PhD, a professor in the Australian Centre for Health Services and Innovation at Queensland University of Technology, described as the “publish or perish” phenomenon.
“If I could do one simple thing tomorrow, I would ban all the university league tables,” said Barnett. “They’re just encouraging corruption.” Ranking systems that prioritize publication volume can push researchers toward quantity over quality, making paper mills an easy way to meet expectations.
Furthermore, publication is often not just a measure of success but a requirement for career progression. “For the clients, it’s believed that they need publications that they can’t achieve through their own efforts, either because they don’t have the time, the facilities, the training, or the money to do research and yet, for whatever reason, their employers expect them to,” explained Jennifer Byrne, PhD, a professor of molecular oncology and lead of the Publication and Research Integrity in Medical Research group at the University of Sydney.
Jennifer Byrne, PhD Professor University of Sydne
Byrne has published extensively about paper mills and publication integrity; she got into the field accidentally when she came across some papers about a gene that her team discovered many years earlier. “In 2014–2015, we realized that five or six different groups suddenly published very similar papers about this gene in different journals,” she said. “And I just thought, that doesn’t really make a lot of sense.”
Upon investigation, Byrne found that the papers, and a further 48 similar publications, showed features consistent with mass production. She has since proposed that human gene research in general is highly vulnerable to paper mills. “You can hide fake research quite effectively in experimental fields, because it’s very difficult and time-consuming to reproduce experimental studies,” she said.
Why paper mills matter
Aside from the obvious fraud, paper mills are problematic for several reasons. Byrne describes them as “a billion-dollar problem” with few resources devoted to tackling it. And although she and others have advocated for scaled investments, progress so far has been slow.
The publishing system can also reinforce the problem. Paper mills are profit-driven, but journals also benefit through article processing charges and citations, creating what Byrne describes as a “circle” in which “everyone gets what they want.”
The consequences of paper mill papers being published can influence real research. The papers are cited, reused, and built upon, wasting both time and money for all involved.
More broadly, the erosion of trust can drive researchers away from entire fields. In Byrne’s case, she stopped doing preclinical cancer research. “I left because there were a lot of papers that I couldn’t trust. When you get to the point where you can’t trust most of the recent literature, it’s very difficult to continue,” she said.
There are also more sinister risks. Barnett recalled reports of paper mills exploiting their clients, including instances of potential blackmail. “If you’ve been a regular customer and then you suddenly stop, they might try and squeeze more money,” he said. “They’ve got absolutely no scruples.”
Despite these impacts, deterrents are limited. “There are almost none,” said Byrne.
Retractions are often slow, meaning damage is done before action is taken, and retraction rates are far below where they should be.
In an April 2026 report to a U.S. Congress hearing on the state of scientific publishing, Kate Travis, managing editor of Retraction Watch, showed that the retraction rate was around 0.2% in 2025, up from 0.02% in 2016. Yet, the report states that Retraction Watch “are confident that the rate […] should be about two percent—10 times what it is today.”
How to tackle the problem
Concerns about problematic papers are often raised by individual researchers or so-called science sleuths on platforms such as PubPeer. Although they have become skilled at spotting telltale signs of a paper mill, like manipulated images, distinct layouts, author affiliations that might not match the topic of the paper, unusual patterns of coauthors, and fake peer reviews, it is difficult for the untrained eye to detect problems from a single paper.
This is why there have been calls for increased awareness. “Awareness is always the first step,” said Byrne, who is working with The Lancet–World Conferences on Research Integrity Foundation commission to address critical issues related to research integrity.
Efforts to extend awareness are also being coordinated through initiatives such as United2Act, which brings together stakeholders from research institutions, publishers, sleuths, and universities to develop shared guidance and educational resources.
But even with greater coordination, human detection has limits. As paper mills scale, automated tools are becoming essential.
Earlier this year, Barnett, Byrne, and colleagues published a paper in the BMJ showing that their large language model (LLM) could flag papers suspected of being from paper mills by analyzing sentence-level patterns. The model identified 9.9% of more than 2.6 million cancer research papers for further review. Many of the papers were linked to regions with strong publication incentives, including China.
However, Barnett emphasized that the model “is not a 100% proof, it’s a quick and simple flag that should encourage reviewers to look at those papers and look for other signs of paper mill activity.”
Other paper mill detection technologies are also available. Platforms such as Clear Skies, which is used by the STM Integrity Hub, use machine learning to detect patterns across large bodies of literature, while image-forensics tools and cross-publisher data sharing help identify duplicated figures and submissions.
Alongside these tools, Barnett suggested that researchers may increasingly need to provide a “breadcrumb trail,” through preregistration of hypotheses and transparent workflows to demonstrate the authenticity of their work.
Platforms such as PubPeer and Retraction Watch also play a role, enabling researchers to flag concerns and share evidence about suspect papers after publication. These flags then prompt journal retractions and investigations, making it a critical component in the fight against paper mill activity.
A call for tighter regulation
Aside from technology, Byrne would like to see tighter regulation for the commercial publishing industry, akin to something like the ISO 9001 quality management standards that have been widely adopted across industries like manufacturing, engineering, and healthcare.
“We need a regulatory framework that rewards journals that do the right thing and that care about publishing quality,” she said. “And we need to disincentivize the current commercial drive towards publishing anything for money.”
Byrne believes that funders and researchers should be demanding these standards. “They pay for the research, the journal subscriptions, the article processing charges, and give their research for free,” she said. “They don’t ask anything in return, in terms of quality standards, and that’s unacceptable.”
Marie Soulière, PhD Elected Trustee COPE
Marie Soulière, PhD, an elected trustee of COPE and chair of the COPE Paper mill Working Group, acknowledged that “a standard such as ISO 9001 could help with process consistency, documentation, and accountability.” But she said, “it would not be a direct solution to publication fraud or paper mills” and “would need to sit alongside integrity-specific controls, not replace them.”
How publishers are responding
Publishers are increasingly shifting from isolated responses to coordinated action. Initiatives like the STM Integrity Hub and United2Act are driving cross-industry collaboration and shared detection approaches.
Soulière said that several recommendations from the COPE/STM 2022 “have been put into practice, particularly around cross-publisher collaboration, shared screening approaches, and investment in integrity infrastructure.”
A central strategy, highlighted in a publication from the United2Act working groups, uses the “Swiss Cheese Model,” a move toward layered screening that combines tools such as plagiarism screening, image forensics, citation analysis, and author verification. “Each safeguard has limitations, but multiple checks together make it harder for fraudulent papers to pass through,” said Soulière.
Adya Misra, PhD Associate Director Sage
Publishers are also strengthening internal processes. As Adya Misra, PhD, associate director of research integrity at Sage, described: “Our research integrity team acts centrally to support editors and internal journal teams with both prevention of suspicious or problematic research and the correction of the scholarly record … in line with COPE guidance.”
A spokesperson for Taylor & Francis highlighted their work on external collaborations designed to address the root causes of integrity issues. They are partnering with the National Science Library at the Chinese Academy of Sciences to develop research integrity and publishing ethics training programs, designed to ensure that students and researchers at all levels receive adequate support and to help them avoid exploitation by unethical third-party services such as paper mills.
AI changes the game
Even as safeguards improve, artificial intelligence (AI) is moving the goalposts. Many current detection strategies were developed to target structured forms of fraud; template-driven papers, recycled images, and repeated patterns across manuscripts. But these signals are beginning to disappear. “Our system worked because the paper mills would have a template, but now with AI, there is no template,” said Barnett. “It’s going to absolutely change everything.”
Barnett and his colleague Matt Spick, PhD, a lecturer in health and biomedical data analytics at the University of Surrey, recently demonstrated this by generating a complete scientific paper in just under 30 minutes using publicly available data and the OpenAI platform PRISM.
“All we did was give it the dataset and said write a paper for an Elsevier journal,” Barnett explained. “If an honors student had given me this paper, I would have been pretty pleased.”
Credit: Hispanolistic / Getty Images
Paradoxically, AI could also be bad news for paper mills as people realize they can create the papers themselves at little to no cost.
Reasons for cautious optimism
With AI adding to the challenges that publishers and researchers already face, the future could appear bleak. Barnett recalled an analogy describing the AI problem as an oil spill in a digital ocean, “We don’t know how deep it is, can’t get to the bottom of it, and it’s very difficult to clean up.”
Even removing a single problematic paper can require significant time and effort, while thousands more remain undetected. But Byrne remains positive that the work being done can have an impact.
“I’m actually really positive, because I think the biggest thing is awareness,” she said, noting that when she gives talks, she asks if the audience has heard of paper mills. “In 2023, that might have been five percent of people, and yet by 2025 it had increased to 30%–50%,” she said.
Soulière added that increased collaboration and transparency within scholarly publishing is another positive takeaway.
“Publishers, editors, institutions, and other stakeholders are no longer treating these issues as isolated problems,” she said. “They are investing in stronger screening systems, clearer policies, and better cross-sector coordination. In that sense, this moment is also driving progress and innovation.
“While the risks are serious, the response from the sector shows that trust can be reinforced, and that the system is becoming better equipped to detect problems earlier and protect the scholarly record more effectively,” Soulière concluded.
Laura Cowen is a freelance medical journalist who has been covering healthcare news for over 10 years. Her main specialties are oncology and diabetes, but she has written about subjects ranging from cardiology to ophthalmology and is particularly interested in infectious diseases and public health.
ObjectiveImplantable sacral anterior root stimulators enable bladder emptying after spinal cord injury but do not prevent reflex incontinence. A closed-loop neuroprosthesis that detects and inhibits reflex bladder contractions could address this, but first, reliable detection of bladder fullness from the sacral roots. Further, the distribution of afferent bladder activity between sacral roots, and the relationship between efferent and afferent activity within each root, remains unclear and must be clarified to guide implant design.MethodsElectrode books were implanted on the S1–S3 extra-dural sacral roots bilaterally in six terminally anesthetized sheep. Afferent electroneurogram (ENG) was recorded concurrently from all implanted roots during filling cystometries and correlated with bladder pressure. Each root was individually electrically stimulated and the bladder pressure response recorded. Post-mortem morphometric analysis determined fiber size distribution in each root.ResultsOverall, S2 ENG activity showed the highest correlation with bladder pressure, and electrical stimulation of S2 and S3 produced the greatest increases in bladder pressure. Fiber size distribution did not correlate with either ENG activity or bladder pressure response. Significant variation was identified between individual sheep, but notably, in four of six sheep, a single sacral root had both the highest ENG correlation to bladder pressure and the greatest bladder response to stimulation.SignificanceThis study demonstrates reliable recording of bladder afferents from sacral roots using clinically applicable electrodes. It provides the first systematic recording of bladder ENG concurrently across three pairs of sacral roots in multiple animals, and the first characterization of signal distribution between roots. Significant individual variation is identified, impacting the design of future implantable sacral neuroprostheses for bladder control.
