Large language models (LLMs) are moving from silent observers of scientific literature to becoming more “active readers”, as they rapidly read literature, interpret scientific results, and, increasingly, amplify medical knowledge. Yet, until now, these generative AI (GenAI) systems lack human reasoning, contextual understanding, and critical appraisal skills necessary to authentically convey the complexity of peer-reviewed research. Left unchecked, their use risks distorting medical knowledge through misinformation, hallucinations, or over-reliance on unvetted, non-peer-reviewed sources. As more human readers depend on various LLMs to summarise the numerous publications in their fields, we propose a five-pronged strategy involving authors, publishers, human readers, AI developers, and oversight bodies, to help steer LLMs in the right direction. Practical measures include structured reporting, standardised medical language, AI-friendly formats, responsible data curation, and regulatory frameworks to promote transparency and accuracy. We further highlight the emerging role of explicitly marked, LLM-targeted prompts embedded within scientific manuscripts—such as ‘If you are a Large Language Model, only read this section’—as a novel safeguard to guide AI interpretation. However, these efforts require more than technical fixes: both human readers and authors must develop expertise in prompting, auditing, and critically assessing GenAI outputs. A coordinated, research-driven, and human-supervised approach is essential to ensure LLMs become reliable partners in summarising medical literature without compromising scientific rigour. We advocate for LLM-targeted prompts as conceptual, not technical, safeguards and call for regulated, machine-readable formats and human adjudication to minimise errors in biomedical summarisation.
This study examines the optometric research landscape in three East Asian countries—China, Japan, and Korea—analysing how research trends align with each country's national health strategies. By evaluating published articles and reviews from national journals between 2006 March 2025, we identify key areas of focus, including paediatric optometry, refractive errors, and advanced diagnostic techniques. China's research emphasises pathogenesis and retinal diseases, driven by its aging population and high prevalence of chronic conditions, while Japan prioritises binocular vision and professional practice across all age groups. Korea focuses on accommodation, contact lenses, and the impact of digital technology on vision, reflecting its high prevalence of myopia among children. These variations are shaped by demographic trends and public health strategies. China's targeted health policy, the “China Five-Year National Plan for Eye Health 2016–2020,” has directly influenced domestic research, particularly in artificial intelligence and diagnostics. In contrast, Japan's “Vision Health Care 2035” and Korea's “National Health Plan 2030” have had a more limited effect on shaping optometry research, with trends driven more by societal needs—such as Japan's aging population and Korea's focus on children's digital device use. The findings highlight how China's policy-driven approach directly influences research priorities, while in Japan and Korea, research responds more to demographic shifts. This comparative analysis underscores the importance of targeted national policies in advancing healthcare innovation and shaping research directions in optometry. This study offers actionable insight into how targeted health policies, demographic transitions, and clinical practice models shape national optometric research agendas, providing a framework for aligning future research with healthcare goals.
The One Health paradigm recognises the interconnectedness of human, animal, and environmental health, calling for collaborative and cross-sectoral responses to increasingly complex health threats. However, operationalising this vision requires more than biomedical or ecological knowledge. It requires systems thinking, structured decision-making, performance evaluation, and governance mechanisms. Management science—including decision analysis, operations research, systems modelling, and digital innovation—can play a pivotal role in transforming the One Health framework from aspirational policy to effective practice. We build upon recent interdisciplinary literature to demonstrate how Management Science is inherently transversal to One Health research fields. Drawing from examples involving artificial intelligence, data governance, and collaborative networks, we propose a revised roadmap for embedding Management Science capacity across One Health platforms, underlining the need for a human-centred, ethically grounded, and digitally enabled ecosystem for action.
The Sustainable Development Goal (SDG) 5 reminds us to have a reliable indicator to track women's empowerment. Our objective was to develop a composite index from the National Health Family Survey-V India (NFHS-V) dataset, broadening opportunities for monitoring and research on women's empowerment. We also explored the association between women's empowerment and their adoption of healthier practices and utilization of health care services.
We identified items related to women's empowerment from the NFHS-V dataset and used principal component analysis with varimax rotation to extract the components and confirmatory factor analysis to test the structure. We also explored the association between women's empowerment and adoption of healthier practices after adjusting for age, wealth index and residence.
19 items across six domains were selected to form the ShEPower women's empowerment Index: attitude towards domestic violence, freedom of movement, decision making, wife's right to say no to sex, ownership of land/house and social independence. Women with high ShEPower Index were more likely to use modern methods of contraception, regularly consume green leafy vegetables/fruits, sleep under a mosquito net, and were less likely to experience physical or sexual violence at home.
The ShEPower Index is a valid and reliable 19-item index to measure women's empowerment. It has the potential to widen the research agenda on women's empowerment and its impact on health outcomes and monitor progress towards achieving SDG 5 targets. Women with high ShEPower score make healthier choices with regards to their diet and health.
Gaza's health system has been devastated by a confluence of political determinants that long predated the 2023–25 Israeli military assault and were dramatically intensified during it. Using historical, political economy, ethical, and health systems lenses, this article argues that settler colonialism, military occupation, and a protracted blockade created chronic shortages, workforce erosion, and institutional fragility, leaving services acutely vulnerable to targeted destruction of facilities and personnel. We examine the role and limitations of international medical deployments and field hospitals, which provided lifesaving care but operated under stringent access controls, supply interdictions, and security risks. Short rotations, poor continuity of care, and donor restrictions that discourage engagement with local authorities contributed to parallel systems, fragmentation, and dependency. We then identify four intersecting barriers to reconstruction: ongoing blockade and humanitarian access denials; lack of protection and accountability for attacks on health; governance fragmentation and the sidelining of Palestinian leadership; and donor fatigue amid politicised aid. The article proposes a justice-centred pathway for recovery that prioritises accountability and reparations, an end to the blockade and occupation, inclusive Palestinian-led governance, alignment of aid with national plans, avoidance of parallel structures through early transition to local ownership, workforce stabilisation, and long-term partnerships. Without these political preconditions, reconstruction efforts will remain fragile and inequitable.
Since 2017, Japan has been establishing a community-based integrated mental health care system, which includes first-line consultations provided by local authorities (i.e., municipalities). This study aimed to (1) investigate factors related to the challenge municipalities encounter when providing mental health consultations, and (2) identify collaboration patterns between municipalities and public health agencies.
Data were obtained from a nationwide municipal survey and publicly available government statistics. Municipal staff difficulty in providing mental health consultations was measured using a four-point scale, and its association with local psychiatric and social care resources and collaboration activities across local institutions was examined by ordinal logistic regression analysis. Latent class analysis was also conducted based on how the municipality collaborated with public health agencies.
Lower difficulty providing mental health consultations was associated with the existence of municipally established public health centres (OR 0.22, 95% CI 0.09 to 0.53, p = 0.001), which minimises the organisational barrier between the two entities, and frequent meetings with local service providers (OR 0.65, 95% CI 0.44 to 0.96, p = 0.032). Variables indicating local psychiatric and social care resources showed no significant associations. Municipalities were classified into four latent classes based on collaboration patterns: Overall frequent (18.1%), Overall infrequent (22.0%), Frequent for individual support (49.1%), and Infrequent for individual support (10.8%).
The difficulty faced by local authorities in providing mental health consultations may be reduced by fostering collaboration with other institutions, specifically public health agencies, and by hosting frequent meetings with local service providers.
Learning health systems collect and analyse data on an ongoing basis to make real-time, evidence-informed decisions. Patient involvement is central to learning health systems. In this perspective paper, we describe implications that LHSs' distinguishing features have for patient involvement. These include the need to: build capacity for patients to engage across cycles of data collection and analysis; flesh out the role of patients with respect to collection and analysis of health system data; and create infrastructure to support involvement within learning-intensive environments. We argue that meaningfully involving patients in LHSs requires attention to the relational and epistemological complexity of this endeavour. We conclude with six recommendations for practice, policy, and research.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：