While developing an ultrasensitive test for the detection of Mycobacterium tuberculosis DNA (TB-DNA), researchers from Boston University have unexpectedly found a high prevalence of the molecular marker in U.S.-born patients hospitalized in Boston.
“We began this research with the intent of sourcing respiratory samples to support the ongoing development of a new molecular assay for TB,” said Guillermo Madico, MD, PhD, scientist at Boston University’s National Emerging Infectious Diseases Laboratories (NEIDL) and co-inventor of the TOP TB assay. “What we found was completely unexpected. Our ultrasensitive test is detecting Mycobacterium tuberculosis DNA in patients who are unlikely to be diagnosed with TB using current methods. This opens the possibility that there could be thousands of Americans infected with forms of tuberculosis disease that remain hidden from our current diagnostic tools—putting them at risk of developing more serious complications or potentially transmitting the disease to others.”
In 2022, there were over 8000 reported cases of TB in the United States, over 600 TB-related deaths, and an estimated 13 million people with Mycobacterium tuberculosis infection. Although incidence has steadily decreased in the U.S., the rate of decline is too slow to meet the ambitious World Health Organization strategy to end the global TB epidemic by 2035.
One threat to the global elimination goal is a gap in the detection of paucibacillary TB disease—a type of TB characterized by a low concentration of M. tuberculosis bacilli in samples that often results in false negative test results.
To improve detection, Madico and colleagues developed an ultrasensitive molecular assay developed at Boston University called the Totally Optimized PCR (TOP) TB assay, which targets a gene involved in M. tuberculosis cell wall assembly.
During the development process, the researchers conducted three separate clinical studies involving 297 patients from Boston hospitals.
Across the studies, the TOP TB assay detected TB DNA in 12–16% of samples—a rate far higher than expected given Boston’s low TB incidence rate. Of note, most TB DNA-positive patients tested negative on standard TB infection tests (tuberculin skin tests or interferon-gamma release assays), and the researchers hypothesize that the findings “indicate the existence of a paucibacillary form of TB that remains unrecognized and is not detectable using current diagnostic tools.”
During the study, there were three patients diagnosed with acute chest syndrome, a life-threatening complication of sickle cell disease, all of whom tested positive for TB DNA.
The researchers point out in Nature Communicationsthat this “previously unrecognized association” has potential implications for clinical care in the U.S. and many other settings.
“These findings suggest we may be missing a significant burden of TB disease, particularly in older Americans and in patients with certain underlying conditions,” said Edward Jones-López, MD, who co-led the study while at Boston Medical Center and Boston University Chobanian & Avedisian School of Medicine. “Most concerning is the potential association with acute chest syndrome in sickle cell patients. If confirmed and expanded upon in larger studies, this finding could lead to better health outcomes for patients with this potentially life-threatening condition.”
The researchers emphasize that their preliminary findings require confirmation in larger, prospective multicenter studies that include comprehensive clinical, radiological, immunological, and microbiological correlation. However, they argue the evidence warrants immediate dissemination given potential implications for medical care and public health.
Background: Many people relapse after achieving abstinence in substance use disorders. Health care providers may scan the horizon for new technologies to predict response that allow interventions to be targeted rather than routine. Currently, no such predictive technologies are available in the United Kingdom. The Subreal app is available for use in research contexts, but no clinical data specific to the app are yet available. Early health economic modeling can use data from the literature to explore characteristics essential for the new technology to be cost-effective. This information can guide developers in setting performance targets and pricing and estimating potential cost savings and/or cost-effectiveness for health care providers. Objective: This study was supported by a UK industry funding body to explore the potential of digital technologies such as the Subreal app to offer cost savings or cost-effectiveness for health care providers. We explored the threshold price and clinical effectiveness required to deliver cost savings and cost-effectiveness in 2 subpopulations with substance use disorders in a UK setting. Methods: Deterministic models were used to estimate costs per relapse and quality-adjusted life years over 1-, 5-, and 20-year time horizons for people who have achieved abstinence after treatment for alcohol or opioid misuse. The intervention was a digital technology predicting relapse, provided—in addition to standard care—for 1 year post achievement of abstinence. In Subreal, biomarker data are collected daily through the app, and artificial intelligence–enhanced risk assessment flags patients who require additional support. The comparator was event-driven, reactive response to relapse. Costs and quality-of-life estimates were calculated using Markov models with data from existing published sources. The base-case estimate of 15% reduction in first-year relapse rates was based on a previous study on a similar but simpler digital technology. Results: Digital technologies such as the Subreal app have the potential to be cost-saving from a UK health and social care perspective, especially when used over a longer time horizon. Assuming a reduction of 15% in first-year relapse rates, digital technologies have the potential to be cost-saving, provided that they do not cost more than £300 (US $400.09) and £460 (US $613.47) per patient per annum for alcohol and opioid use disorders, respectively. No cost was included for postalert care, as it was assumed that this could be met within existing resources. Cost savings would be achieved predominantly through a reduction in treatment requirements as fewer people relapse. Price thresholds would reduce correspondingly if a <15% reduction in relapse rates were achieved. Conclusions: Developers of digital technologies that aim to reduce relapse need to focus on the generation of evidence of clinical effectiveness and develop a commercially sustainable pricing model that allows health care providers to benefit from cost savings.
AWS has now unveiled Amazon Bio Discovery, an AI platform that grants researchers direct access to a broad library of biological foundation models that can be fine-tuned for specific use cases in drug discovery. Announced at the AWS Life Sciences Symposium at the Javits Center in New York, the platform is supported by an AI agent that can select models for research goals, and evaluate candidates for synthesis and testing to enable a rapid lab-in-the-loop experimentation cycle.
While rising AI models show promise, they require coding skills and the ability to manage computing infrastructure. Additionally, diverse models face benchmarking challenges and moving candidates from computational design to physical synthesis remains a multi-step process. Given that data live in disconnected systems, scientists must manage multiple lab partners and manually coordinate timelines and execution.
Amazon Bio Discovery addresses these challenges with three capabilities: a benchmarked library of AI models and analysis packages, an AI agent that supports experimental lab, and integrated lab partners that test top antibody candidates and route results back to the researchers. This feedback loop improves the next round of design.
“AI agents make powerful scientific capabilities accessible to all drug researchers, not just those with computational expertise,” said Rajiv Chopra, PhD, vice president of AWS Healthcare AI and Life Sciences.
Currently, 19 of the top 20 global pharmaceutical companies use AWS to power research workloads. Amazon Bio Discovery will bring enterprise-grade scale, privacy, and security to researchers across pharmaceutical, biotech, and academic research organizations. MSK, Bayer, the Broad Institute, and Voyager Therapeutics are among early adopters of Amazon Bio Discovery.
Among the Amazon Bio Discovery broad catalog includes open-source and commercial models from Apheris and Boltz. Biohub and Profluent are expected to join the platform.
Amazon Bio Discovery enables scientists to fine-tune the model by feeding prior experimental data from their organization’s lab results into the application without complex training pipelines or custom code. In-house models can also easily be deployed and hosted within Amazon Bio Discovery.
To support model selection, an antibody benchmark dataset is available to evaluate the likelihood of a drug candidate to have favorable biological properties, such as manufacturability and stability.
Candidates selected for experimental validation can be directly sent to Amazon Bio Discovery’s integrated network of laboratory partners, including Twist Bioscience, Ginkgo Bioworks. A-Alpha Bio is also anticipated to join the network.
Gene editing has been able to silence the extra chromosome that is mostly responsible for Down syndrome (DS) in a cell-based study that could be the first step towards therapeutic treatment.
The approach uses a modified form of CRISPR-Cas9 to alter precise sections of DNA.
Researchers used this to insert the X-inactivation specific transcript (XIST) gene to deactivate X chromosomes in female cells, thereby correcting chromosomal triplication.
The partial transcriptional correction, reported in PNAS, offers a scalable, targeted platform for chromosomal therapy in Down syndrome and other aneuploidy disorders, which are conditions involving an abnormal number of chromosomes.
“These studies overcome a major hurdle in the treatment of Down syndrome (a genetic disorder cause by three rather than two copies of chromosome 21) by dramatically increasing the efficiency with which to insert XIST onto a single copy of chromosome 21 and thereby silencing this third copy,” senior researcher Volney Sheen, PhD, from Beth Israel Deaconess Medical Center, told Inside Precision Medicine.
“Determination of the best ways and times to deliver these constructs to the brain will become the next focus as we seek a clinical treatment for DS.”
Down syndrome is the most common genetic disorder and occurs in one out of 700 live births. It is linked with cognitive impairment, heart defects, and early-onset Alzheimer’s disease and results from the triplication of approximately 500 genes as well as other genetic changes on chromosome 21.
The XIST gene produces a long, noncoding RNA that inactivates many of the genes on one of the two X chromosomes of female mammals. It has therefore been mooted as a treatment for Down syndrome but technical limitations, including low levels of gene integration, have hindered progress.
In an attempt to address this, Sheen and team created a CRISPR-based method that involved fusing a codon-optimized λ-phage with Cas9, assembling single guide (sg)RNAs specific to single nucleotide polymorphisms (SNPs), and enhancing donor-acceptor DNA pairing.
The modified CRISPR-Cas9 method achieved a high level of genomic integration of large genetic material, improving the efficiency and specificity of XIST integration.
Inserting XIST onto one of the trisomic chromosome 21 alleles using SNP-dependent sgRNAs achieved an integration efficiency of 20% to 40% for the 14kb XIST gene.
XIST integration was revealed through expression of the enhanced green fluorescent protein reporter, clonal sequencing of individual lines, and fluorescent in situ hybridization.
The team further demonstrated that XIST activation led to upregulation of epigenetic markers, broad downregulation of messenger RNA expression on chromosome 21, and downregulation of specific genes on this chromosome.
“Our findings demonstrate atrial transcriptional correction of trisomic gene dosage and offer a scalable targeted platform for chromosomal therapy in DS and other aneuploidies,” the researchers reported.
They added: “The modified CRISPR method with XIST paves a rode for therapeutic treatment for DS.”
Twenty years ago, Sven Bocklandt, PhD, sought to create a hypoallergenic cat. He had the genetic engineering chops to do it, but the embryology was beyond his capabilities. At a small animal genetic engineering conference, known as TARC (Transgenic Animal Research Conference), held near Lake Tahoe, he met James Kehler, VMD, PhD, whose research at that time was to make transgenic and knockout cats as models of human disease.
The two men bonded, agreed the hypoallergenic cat idea was “crazy enough,” and decided to move forward with it. They worked together, completely unfunded, for years—FedEx’ing samples back and forth as Bocklandt was on the west coast and Kehler on the east coast—trying to make their “garage cat” while each one worked different day jobs.
Bocklandt, passionate about animal genome engineering, continued to develop different ideas for genome engineering in animals. Around the same time that he started sharing his ideas with scientists like George Church, PhD, a start-up focused on animal genome engineering was taking shape—Colossal Biosciences, co-founded by Church. Introductions were made, and Bocklandt joined in 2022 as species director to work on the dire wolf project. Kehler joined a short time later as VP. And everyone knows the rest of that story (there was no shortage of media coverage).
The pair eventually succeeded with the cat project: his name is Archie, and he is, Kehler noted, only partially hypoallergenic. But the generation of Archie and the dire wolves may not be the successes of this story. The real success may be what Bocklandt and Kehler learned along the way—and what they are going to do next.
Chickens as the next biologic factory
Neion Bio, co-founded by Dimi Kellari and Sam Levin, PhD, and located on the Rockefeller University campus on the east side of Manhattan, is aiming to re-engineer eggs to produce drugs in chickens. The team uses genetic engineering to integrate therapeutic proteins into native egg proteins, creating a new manufacturing platform for drugs that runs on grain and water.
Bocklandt joined the team at Neion Bio as CSO after leaving Colossal in 2024; Kehler joined more recently, as head of avian sciences.
When thinking about producing complex proteins, using the chicken “makes a lot of sense,” Bocklandt told GEN. Breeding and genetic engineering are all established in the chicken. And the vaccine industry has established an existing infrastructure to grow eggs under disease-free conditions. Purifying proteins out of an egg, Bocklandt added, is easier than purifying them out of a Chinese hamster ovary (CHO) culture (the traditional cell choice for drug production) because there are fewer host proteins.
Sven Bocklandt, PhD [Marco Figueroa]
It makes “far more sense” than what we’re doing right now, Bocklandt noted, which is using CHO cells. “Everyone is doing that because everyone has been doing it that way,” he asserted.
“The fact that we’re now seriously questioning whether CHO cells should remain the default manufacturing platform for biologics is long overdue,” noted Ola Wlodek, PhD, CEO of Constructive Bio. “Any credible new approach that breaks this decades-old lock-in is ultimately good for patients and for the field.”
For Kehler, who did his graduate work in the lab of stem cell pioneer Hans Schöler, PhD, the chicken is a clear choice because it is the only species, besides the mouse, where the primordial germ cells have been used to transmit genetically modified gametes to the next generation.
Mike McGrew, PhD, group leader at the Roslin Institute in the U.K., and an advisor to Neion Bio, demonstrated years ago that modifying chicken primordial germ cells is a reliable way of making gene-edited chickens. This background is comforting to Kehler, who noted that, “unlike at Colossal, where everything was bleeding edge, we are able to focus on a single species and capitalize on some pretty tried and true technology.”
Drugs in eggs meet biomanufacturing reality
The lab space on the Rockefeller University campus can support research and even house chickens. But it cannot support the production of a drug. When asked about turning their egg-borne proteins into drugs, the company leans on the existing infrastructure that supports vaccines in specific pathogen free (SPF) eggs. The idea is that the egg whites will be frozen in giant batches and then processed in a CDMO.
When asked about potential challenges, Bocklandt noted that, “technically, there’s not much to worry about. I have no concerns about Neion Bio being able to do what we want to do or what we need to do.”
But there may be hurdles ahead. Rahul Dhanda, co-founder, president, and CEO of Syntis Bio, told GEN that “at the beginning, everything can look like it has infinite potential—it’s when you actually build and operate the system that the real challenges show up.”
More specifically, Dhanda pointed out that biomanufacturing “ultimately comes down to reliable, consistent, and cost-efficient production.” Leveraging animal biology for drug manufacturing is exciting, he noted, “but scalability and cost are still open questions, especially at this early stage. Biological variability between animals and individual outputs, like eggs, introduces additional risk compared to more controlled cell-based systems,” Dhanda added.
Wlodek agreed: “because egg-based production is inherently a biological supply chain, it will face avian flu risks, batch-to-batch variability from seasonal and flock effects, animal-welfare/regulatory overhead, and practical limits on how fast you can expand output compared with stainless-steel or single-use fermenters.”
Microbial and yeast systems still “win decisively on GMP containment, land/water footprint,” she noted, and “the ability to go from a few liters to tens of thousands of liters in weeks rather than months.”
Dhanda agreed that “getting it to work in principle is far different from getting it to work at scale, and that seems far off.”
If these challenges can be addressed at scale, safely and humanely, Dhanda noted, the approach could deliver meaningful health benefits—”but there are still significant logistical and technical hurdles to work through.”
Engineering the chicken genome
Creating dire wolves at Colossal started with deriving wolf cells, editing them, and cloning them back into a live animal. But cloning doesn’t exist in birds. To genetically engineer chickens, the Neion Bio team edits the germline, starting the process with a fertilized egg.
Neion Bio [Marco Figueroa]
The egg is incubated for 65 hours, at which point germ cells float in the blood because the ovaries and testes don’t exist yet. A microliter of the blood is removed, put into cell culture media, and the germ cells grow out. The transgene that codes for the therapeutic protein is inserted using CRISPR-Cas enzymes, in the coding region of a gene that codes for Ovalbumin—which makes up a bit over 50% of the egg white protein. This protein is made “on a massive scale” by the oviduct, the company noted.
The genome is screened for correct integration and potential off-target edits. Once the clone is approved, several thousand cells are injected back into another embryo (also at 65 days old). After incubation, the egg hatches and becomes a chicken.
Kanuma set the precedent—but not the scale
In 2015, the U.S. Food and Drug Administration approved Kanuma (sebelipase alfa) to treat Lysosomal Acid Lipase (LAL) deficiency, also known as Wolman disease. The drug, an enzyme replacement therapy, was the first treatment for patients with the rare disease and the first drug to be made in chickens. Kanuma is produced by Alexion Pharmaceuticals, which was acquired by AstraZeneca in 2021.
This historical precedent may provide a proof of concept for Neion Bio. That said, “the scale required for Kanuma is very different from what would be needed for large biosimilars,” explained Wlodek.
An Odyssean journey
For both Bocklandt and Kehler, the move to Neion Bio feels like their careers are coming full circle. When Bocklandt first left Colossal, he was not sure how he would surpass that level of excitement. But the move came at an interesting time for him; the call to join Neion Bio came just weeks after he learned that his sister had been diagnosed with leukemia.
He thought, “Well, maybe this is not such a bad use of my skills.”
Earlier in his career, he didn’t think that he had anything special to add to a field like cancer research. But now Bocklandt sees it differently: throughout his career, he has pushed the state-of-the-art of genetic engineering. Now, he said, “I bring something to the field. And the fact that I can do my passion, animal genetic engineering, and apply that to make drugs better, cheaper, and more accessible, is really exciting.”
As for Kehler, Neion’s goal was his goal all along. He went to the University of Pennsylvania to make better animal models to test drugs for humans. “It never really dawned on me that we could use animals to make the drugs for humans. But taking everything I know about stem cell biology, germ cell biology, and gene editing, and bringing that to bear to make what should be a disruptive, transformational approach to making drugs—it feels like the culmination of my career.”
Neion (pronounced Neon) Bio is named after the birthplace of Odysseus; Mount Neion is a mountain mentioned in Homer’s The Odyssey as a landmark on Ithaca—Odysseus’ island home. As described by the company, the name is a testament to the shared qualities between the Greek hero and the company’s goals: relying on intelligence and resourcefulness over strength. And yes, Odysseus was successful in his return home to reclaim his throne. But it was a bittersweet success given the enormous cost and hardship.
Neion Bio’s name may mirror the resilience and ingenuity required to undertake the journey, but time will tell how long the similarities in the namesake are shared between the two.
Rise and shine, everyone, another busy day is on the way. And it is getting off to a good start here on the Pharmalot campus, where clear blue skies and unusually balmy breezes are greeting us. Who could ask for anything more? Actually, we could — it is time to reheat the kettle for another cuppa stimulation. Our choice today is elderberry-infused green tea. And here is a helpful tip — a teaspoon of honey enhances the flavors splendidly. Of course, you are invited to join us. For the full experience, we are now hawking replicas — take a look. Meanwhile, here are a few tidbits to help you along. As always, do keep in touch. We appreciate feedback, suggestions, criticism, and tips. …
In a bid toward greater transparency, the U.S. Food and Drug Administration sent reminder letters to more than 2,200 companies and researchers that they are required to report clinical trial results to a federal government database or they may face fines, STAT says. FDA officials disclosed that an internal analysis found results were not submitted for nearly 30% of studies that were “highly likely” to fall under mandatory reporting requirements. The agency also noted that the letters were sent to companies and researchers associated with more than 3,000 registered trials, some of which were publicly funded. In explaining its move, the regulator acknowledged a long-standing complaint from researchers who have argued that without access to specific data, trial results cannot be easily duplicated, which inhibits greater understanding of how medicines might work.
Novo Nordisk and ChatGPT maker OpenAI agreed to work together on how to leverage artificial intelligence to discover new drugs, the latest AI partnership in the medical field as health care companies seek to harness the technology to get ahead of the competition, The Wall Street Journal tells us. The drugmaker said it would integrate OpenAI’s models across its operations to help its workforce analyze complex datasets and reduce the time it takes to move from research to delivering treatments to patients. The group said the partnership would boost efficiency across the organization, with pilot programs initially launching in research and development, manufacturing, and commercial operations ahead of a full AI integration by the end of the year.
IntroductionMultiple sclerosis (MS) affects different cognitive domains, including social cognition. Immersive Virtual Reality (VR) may provide a novel rehabilitative approach to treat motor and cognitive symptoms of MS. This exploratory pilot study evaluated the effects of immersive VR rehabilitation on social cognition in MS patients and explored related cortical neurophysiological signatures.MethodsSeven MS patients underwent immersive VR rehabilitation with the CAREN system (3 sessions/week, approximately 45 min of active training per session, about 1 h including preparation, 8 weeks), while seven healthy controls (HC) did not undergo any intervention. Patients were evaluated at baseline (T0) and post-treatment (T1) with standardized measures of cognitive, emotional, and motor functioning. EEG data were acquired from all participants, and, after artifact removal, spectral parameterization decomposed signals into aperiodic (exponent, offset) and periodic oscillatory components (alpha and beta power). Power spectral density was analyzed using group comparisons and Pearson correlations with neuropsychological measures.ResultsCompared with HC, MS patients showed reduced alpha-band power, mainly over frontal and parieto-occipital regions, whereas aperiodic parameters did not differ between groups. In patients, alpha and beta power correlated with the Positive Emotions Self-Efficacy Scale (alpha: r = 0.92, p = 0.003; beta: r = 0.83, p = 0.020). Alpha power is also correlated with RAO SRT–LTS (r = 0.85, p = 0.016), and beta with EQ-CE (r = 0.82, p = 0.023). Overall, alpha and beta power were correlated with emotional self-efficacy, balance, memory, and empathy, suggesting that oscillatory markers are potential indicators of clinical outcomes.DiscussionRehabilitation via immersive VR has shown promising clinically significant effects in the cognitive, emotional, and motor domains, supported by convergent EEG spectral signatures. Future studies employing predictive modeling approaches will be required to assess their prognostic value.
IntroductionDeep learning approaches have become central to brain MRI analysis; however, their reliability under dataset shift remains a critical barrier to safe and scalable deployment in neuroscience and clinical research. While convolutional neural networks (CNNs) provide strong locality-driven inductive biases for robust feature extraction, they lack global contextual awareness. Conversely, transformer-based architectures capture long-range dependencies but often exhibit reduced robustness and miscalibrated confidence when applied to heterogeneous medical imaging data, particularly in Cross-Dataset settings.MethodsIn this work, we propose a calibration-aware hierarchical CNN-Transformer fusion framework designed for robust brain MRI analysis under dataset shift. The architecture integrates a pretrained multi-scale CNN backbone with a hierarchical transformer branch and performs scale-aligned fusion through cross-attention mechanisms. By allowing local convolutional features to selectively query global contextual representations, the proposed design maintains stable feature contributions during fusion and mitigates overconfident reliance on transformer features when generalization degrades across datasets. The framework is evaluated using a strict Cross-Dataset protocol, where models are trained on one dataset and tested on a distinct dataset.ResultsExperimental results demonstrate that the proposed fusion model achieves competitive classification performance while substantially improving probabilistic calibration relative to both CNN-only and transformer-only baselines. Specifically, the model attains an average accuracy of 99.20% and achieves lower Expected Calibration Error (ECE = 0.0041), Brier score (0.0028), and Negative Log-Likelihood (NLL = 0.0277) compared to a standalone Swin Transformer and a strong ResNet50 baseline.DiscussionThese findings demonstrate that calibration-aware hierarchical CNN-Transformer fusion enhances both predictive reliability and robustness under Cross-Dataset evaluation. By improving the alignment between predictive confidence and empirical correctness, the proposed method supports safer large-scale analysis of heterogeneous brain MRI data, with important implications for multi-center neuroscience studies and trustworthy clinical decision support.
IntroductionRestrictive interventions, including physical restraint, seclusion, chemical restraint, and segregation, continue to be used within mental health services, despite sustained policy efforts to promote least-restrictive and trauma-informed care. However, little is known about national trends affecting women, for whom restrictive interventions often carry heightened risks of re-traumatisation and stigma.MethodsWe conducted a longitudinal secondary analysis of publicly available administrative data from the Mental Health Bulletin covering NHS-funded mental health services in England between 2017 and 2025. Annual counts of restrictive interventions involving women were examined relative to the number of women detained under the Mental Health Act to estimate annual rates per 1,000 detained. Regression modelling was used to assess temporal trends overall, by age group and type of restrictive intervention, and interrupted time-series analyses to examine changes following implementation of the Mental Health Units (Use of Force) Act 2018 (“Seni’s Law”). Trends were also examined alongside available national data on restrictive interventions involving men.ResultsRates of restrictive interventions involving women increased by approximately 12 percent per year over the study period, with no evidence of a reduction following the introduction of Seni’s Law. Increases were most pronounced for chemical restraint, seclusion, and segregation, while physical and mechanical restraint remained stable. Restrictive interventions declined among women under 18 but increased consistently across all adult age groups, indicating a widening age-related divergence. Although overall trends broadly mirrored those observed among men, the types of restrictive interventions used and their potential impact may differ, highlighting gendered dimensions in how restrictive practices are experienced and applied.DiscussionDespite extensive national initiatives, restrictive interventions involving women have continued to rise in England, highlighting a persistent gap between policy intent and practice. The findings suggest that legislative frameworks alone are insufficient to achieve meaningful reductions without operational changes in clinical practice, organisational culture, and monitoring systems. Internationally, the study contributes rare gender-disaggregated longitudinal evidence and highlights the need for comparable monitoring systems and coordinated research to inform rights-based, trauma-informed strategies to reduce restrictive interventions in mental health services.
