bims-librar 2026-11-08 papers

bims-librar

Biomed News

on Biomedical librarianship

Issue of 2026–11–08
twenty-six papers selected by
Thomas Krichel, Open Library Society

Embedded Librarians and Library Services Impact Nursing Excellence and the Magnet® Journey.
Protocol for a Crowd-Sourced Living Global Directory of Grey Literature Sources for Systematic Reviews of Evaluations of Climate Solutions.
Prompting Strategies for Readability Alignment Between User-Generated Ophthalmology Questions and Large Language Model Answers: Comparative Evaluation Study.
An Agentic AI Pipeline for an Evidence-Grounded Environmental Health Public Information System: Development and Evaluation.
Readability of Artificial-Intelligence-Generated Responses to Patient Questions About Hearing Loss and Dementia.
Navigating digital surgical advice: evaluating the reliability of large language models for pediatric phimosis.
Clinical accuracy, information quality and readability of chatbot responses to patient questions about lower extremity amputation.
Quality, Readability, and Reliability of Artificial Intelligence Chatbot Responses to Parent-Centered Questions on Newborn Hearing Screening: A Cross-Sectional Comparative Study.
Online Information Available to Caregivers on Introducing Solid Foods to Infants with Cleft Palate: A DISCERN-Based Cross-Sectional Analysis.
Assessing the Readability and Quality of Online Patient Information Regarding Aspirin-Exacerbated Respiratory Disease.
The use of different artificial intelligence-powered chatbots to answer questions related to dental implants.
Readability, understandability, actionability and information quality of online patient information on diagnostic imaging during pregnancy and breastfeeding: A cross-sectional study.
Information on Greek Web Pages About Dietary Weight Loss Should Be More Accurate and Easier to Understand.
An Appraisal of Printed Online Education Materials on Transoral Robotic Surgery (TORS) Online Education for Patients.
Quality and readability of online patient education materials for trigeminal Neuralgia: A cross-sectional review.
Clinical accuracy and applicability of YouTube videos on rheumatic diseases and pregnancy.
Quality and Reliability of YouTube Videos as a Source of Information Regarding Early Signs of Autism Spectrum Disorder.
YouTube Videos on Arthrogenic Muscle Inhibition Show Low-to-Moderate Quality and Reliability.
Quality and reliability of youtube videos on CPAP, BiPAP, and non-invasive ventilation: a cross-sectional content analysis.
Quality and reliability of short videos about type A aortic dissection on Bilibili and TikTok: a cross-sectional study.
Information quality, clinical content, and engagement of robot-assisted nephrectomy videos on Bilibili and Douyin: a cross-sectional study.
Information quality of implant-related videos on Chinese social media platforms: a cross-sectional analysis of Bilibili, Douyin, and Xiaohongshu.
Gynecomastia Trends on TikTok.
Quality and patient-safety content of Instagram videos on ophthalmic emergencies: A cross-sectional content analysis.
Social Media Literacy and Health Information-Seeking Behavior Among School Students and Educators in India: A Needs Assessment Under Project Info-Suraksha.
Evaluating the Performance of Five Leading Large Language Models in Generating Health Education Materials for Patients With Infertility: Comparative Analysis.

J Nurs Adm. 2026 Nov-Dec 01;56(11):56(11): E44-E47

Embedded Librarians and Library Services Impact Nursing Excellence and the Magnet® Journey.

Ryan D Norman, Jeannine Creazzo, Eileen Allen, Jacqueline Ciccarelli.

Two master's-trained hospital librarians, a director, and a medical librarian are employed by a comprehensive healthcare system in the Northeast to provide library services to four hospitals varying in size, services, and populations. Three of the four hospitals have been recognized with the American Nurses Credentialing Center (ANCC) Magnet® designation, and one hospital is pursuing the ANCC Pathway to Excellence Program® designation. The 2 librarians support clinical and nonclinical staff at all 4 hospitals in the conduct of literature searches, securing full-text article requests, managing library database access, offering current awareness services, and designing and running educational classes. The librarians participate as regular members on nursing councils, including those that support Shared Governance, an organizational structure which empowers nurses to take part in shared decision-making at the hospital. The librarians also partner with nursing to maintain a culture of nursing inquiry, practice evidence-based nursing, and strive for nursing excellence. In doing so, the librarians have become an important element of the hospital's Magnet® designation success. Embedded librarians have a positive impact on nurses' utilization of library services and support nurses' evidence-based practice, research, and quality improvement initiatives. There is an opportunity in the future to closely track the number of nurses who use library services, to continue to observe the impact librarians have on the pursuit of nursing excellence, and to research the effect of librarians on Magnet® recognition.

DOI: https://doi.org/10.1097/NNA.0000000000001801
Campbell Syst Rev. 2026 Sep;22(3): 18911803261486366

Protocol for a Crowd-Sourced Living Global Directory of Grey Literature Sources for Systematic Reviews of Evaluations of Climate Solutions.

Paul Fenton Villar, Nicholas Philip Simpson.

  Under intensifying climate risks, national and international policy assessments face growing pressure to engage with rapidly expanding and complex bodies of evidence. Systematic reviews are increasingly occupying a central role in supporting such assessments, using structured and transparent methods of evidence identification and appraisal to consolidate findings from evaluations of the impacts of climate-related policies, programmes, and interventions. However, comprehensive identification of relevant evaluation evidence remains a persistent methodological challenge, particularly where studies are disseminated through grey literature sources that are not indexed in academic databases (such as many government evaluation reports, foundation-commissioned research, and evaluations produced by private and third-sector organisations). Based on expert-developed resources, current guidance and resources to support searching for grey literature is not comprehensive, can become quickly outdated, and does not reflect the full range of potential grey literature sources, instead tending to provide information on more visible and internationally recognised organisations. This research proposes to explore the development of a crowd-sourced, living global directory to inform understanding of the range of different of grey literature sources that exist on evaluations of climate solutions, and how these sources are distributed according to type and geographic context. Based on a community contribution model and the principle of collective knowledge (or the "wisdom of the crowd"), the approach seeks to aggregate knowledge distributed from individuals across the globe, who are embedded in diverse policy and geographic contexts. This recognises that awareness about where evaluation evidence resides can be locally or institutionally held. This protocol outlines the inclusion criteria, identification strategy, and classification framework guiding the development of the directory, as well as the subsequent planned analysis of its findings.

Keywords:  Climate change; Crowdsourcing; Grey literature; Information retrieval; Systematic reviews

DOI:  https://doi.org/10.1177/18911803261486366
JMIR Form Res. 2026 Nov 05. 10 e97408

Prompting Strategies for Readability Alignment Between User-Generated Ophthalmology Questions and Large Language Model Answers: Comparative Evaluation Study.

Mário César Bulla, Daniel Lavinsky, Alexandre Antonio Marques Rosa, Pedro Carlos Carricondo.

   Background: Accessible health communication is essential for effective patient education, yet most medical information remains written at a level above the general population's recommended reading level. Large language models (LLMs) can simplify complex medical information, but it remains unclear whether responses can be automatically adapted to match the linguistic complexity of user-generated questions seeking medical information.
Objective: To evaluate whether the LLM-generated responses to user-generated ophthalmology questions align with their readability and to compare different prompting strategies for readability control across multiple LLM architectures, including their impact on response quality as well as computational cost and efficiency.
Methods: A total of 400 user-generated ophthalmology questions collected from public YouTube comments were categorized into 3 readability tiers based on the Flesch Reading Ease Score. A controlled comparative evaluation of responses generated by 4 LLMs (GPT 5.2, Gemini 3, Claude Opus 4.5, and OSS-20b) was performed under 4 prompting conditions: Baseline (no readability control), Formula-Informed Readability Prompt, Semantic Readability Inference, and External Calculator-Guided Prompt. Agreement between question and response readability was assessed using quadratic-weighted Cohen kappa. Response quality was evaluated using BERTScore and LLM-as-a-judge assessments. Computational efficiency was evaluated through token usage, estimated cost per response, and latency.
Results: The external calculator-guided strategy produced the strongest alignment in readability across models, achieving the highest agreement (Gemini 3: κ=0.94; GPT 5.2: κ=0.90), while maintaining favorable computational efficiency. Formula-informed prompting showed intermediate performance, whereas semantic inference demonstrated poor agreement due to unreliable readability estimation. High-performing models (GPT 5.2 and Gemini 3) maintained very high factual accuracy (>98%) across strategies, indicating that readability control did not substantially degrade correctness. In contrast, the open-source OSS-20b model showed lower accuracy (83.75%-90%) and greater sensitivity to readability constraints. Semantic similarity remained high across models (Bidirectional Encoder Representations from Transformers Score [BERTScore] F1>0.87). Longer questions were associated with slightly lower accuracy, particularly in lower-performing configurations.
Conclusions: Aligning LLM responses to the readability of user-generated questions is feasible and can be reliably achieved using externally guided readability signals. Hybrid pipelines that combine traditional readability metrics with LLM-generated content may therefore represent a practical approach for scalable, readability-adaptive health communication systems.

Keywords:  artificial intelligence; health communication; health literacy; natural language processing; ophthalmology; readability

DOI:  https://doi.org/10.2196/97408
medRxiv. 2026 Oct 29. pii: 2026.10.26.26366549. [Epub ahead of print]

An Agentic AI Pipeline for an Evidence-Grounded Environmental Health Public Information System: Development and Evaluation.

Stein H Ingebretsen, Ram Gouripeddi, Mollie R Cummins, Julio C Facelli, Fatemeh Shah-Mohammadi.

Background: Existing environmental and health information systems typically provide isolated data streams, requiring users to interpret heterogeneous information sources without personalized guidance. Residents and health providers of urban regions with complex environmental conditions, such as the Salt Lake Valley, frequently seek timely, location-specific information about weather, air quality, and their associated health impacts, but may find it difficult to integrate this information in their decision-making, precluding the effective translation of valuable environmental health information into action.
Objective: This study aimed to develop and evaluate an agentic artificial intelligence (AI) pipeline that delivers evidence-grounded, neighborhood-scale responses to health-related environmental questions raised lay persons by dynamically orchestrating authoritative weather, neighborhood-scale environmental observations, air quality, and health-evidence resources.
Methods: We developed an agentic AI pipeline using LangGraph and OpenAI large language models to interpret natural language queries, select and invoke relevant external tools, retrieve real-time environmental data and evidence-based health guidance, and synthesize responses. The pipeline integrates multiple authoritative resources, including weather forecasts, air-quality observations, geospatial services, and a retrieval-augmented generation (RAG) module built from asthma-related clinical guidance. The system incorporates fallback mechanisms to improve robustness when external services are unavailable. We evaluated the pipeline using a curated benchmark of representative user questions, assessing tool-needed recognition, correct tool selection, avoidance of irrelevant tools, parameter mapping, analytic pathway selection, and incorporation of tool outputs into the final response.
Results: The developed system successfully integrated heterogeneous environmental and health information sources into a unified conversational interface that produces evidence-grounded, neighborhood-scale responses for the Salt Lake Valley. Evaluation demonstrated generally strong tool recognition, parameter mapping, and incorporation of retrieved information across diverse question types, supporting the pipeline's ability to retrieve and integrate authoritative environmental and health information consumable by lay persons and providing an important step for the T3 and T4 phases of the translational pipeline of exposure health informatics.
Conclusions: By combining existing environmental health knowledge with local sensor observations and neighborhood-scale environmental measurements, and authoritative state and national weather and air quality resources, the system delivers more localized and context-aware guidance than approaches that rely on authoritative regional or state-level information. Rather than depending on the internal knowledge of a large language model, the pipeline dynamically retrieves, synthesizes, and attributes evidence from authoritative sources. The modular architecture enables seamless integration of additional environmental data streams, health knowledge sources, and geographic regions, providing a scalable foundation for future translational pipelines, completing the T3 to T4 path for environmental health.

DOI: https://doi.org/10.64898/2026.10.26.26366549
Noise Health. 2026 Sep-Oct 01;28(134):28(134): 1360-1367

Readability of Artificial-Intelligence-Generated Responses to Patient Questions About Hearing Loss and Dementia.

Betty Kollia, Corey Basch, Kiara Polanco, Grace C Hillyer.

   OBJECTIVE: This study examines the readability of artificial intelligence (AI)-generated responses to public queries regarding hearing loss (HL), dementia, and the link between them.
METHODS: This cross-sectional study submitted 20 standardized patient-facing questions about HL, dementia, and their relationship, to three AI platforms (ChatGPT, Claude, and Perplexity AI). At the time of data collection, the default models were, respectively, GPT-5.3 Instant, Claude Sonnet 4.6, and Perplexity AI default "best" model selection. All 60 responses were assessed for readability using two validated metrics, Flesch-Kincaid Grade Level (FKGL) and Simple Measure of Gobbledygook (SMOG). The American Medical Association (AMA)/National Institutes of Health (NIH)-recommends eighth-grade reading level for patient education materials. Descriptive statistics were calculated for readability scores across platforms and 10 topic domains (SPSS 31.0).
RESULTS: Across the three AI platforms queried, FKGL median readability for responses to hearing-related questions was at the upper high school to junior college grade level (12.3-15.3) and 1-2 grade levels lower with SMOG (11.4-13.0). For dementia responses, the readability was at 15.2-16.3 grade level with FKGL and 12.9-14.1 using SMOG. No statistically significant differences were found between AI platforms overall and by domain. Claude mostly received high grade level scores for domains regarding symptoms and condition management, for both HL and dementia with FKGL and SMOG.
CONCLUSION: AI chatbots are used by the public to seek health information, yet this study found that ChatGPT, Claude, and Perplexity AI produced responses to questions about HL and dementia at 3-8 grade levels above the AMA/NIH-recommended eighth-grade benchmark. These findings align with a growing literature showing that AI-generated health content routinely exceeds recommended readability thresholds across many medical specialties. This is particularly concerning for the aging populations most affected by HL and dementia. The authors urge relevant professional organizations to audit their public-facing materials for readability compliance.

Keywords:  artificial intelligence; chatbots; dementia; health literacy; hearing loss; large language models; readability

DOI:  https://doi.org/10.4103/nah.nah_159_26
World J Pediatr Surg. 2026 ;9(6): e001276

Navigating digital surgical advice: evaluating the reliability of large language models for pediatric phimosis.

Adrian Chi Heng Fung, John Yeh Joon Park, Eunice Cheuk Lam Lee, Carol Wing Yan Wong, Fanny Yeung, Ivy Hau Yee Chan, Patrick Ho Yu Chung, Kenneth Ky Wong.

   Background: Phimosis is an extremely common pediatric surgical condition. Given the convenience and accessibility brought by artificial intelligence, the readability, quality and accuracy evaluation on the information provided by large language models (LLM) is of utmost importance in providing comprehensive clinical care.
Methods: Six case questions relating to phimosis on its definition, diagnosis, treatment options and potential complications were inputted into each LLM-Chatbots (ChatGPT-3.5, Gemini-1.5 and Claude-3). All responses were reformatted, shuffled and recombined into three passages for single-blind scoring on accuracy by three pediatric surgeons with reference to European Association of Urology (EAU) pediatric urology guidelines. The responses were also assessed for readability using the Gunning-Fog score, Simple Measure of Gobbledygook (SMOG) index and Dale-Chall score and for quality using the DISCERN score.
Results: Comprehension of the information was considered difficult with a mean Gunning-Fog score, SMOG index score and Dale-Chall score of 14.2 ± 2.6, 11.3 ± 2.2 and 9.9 ± 0.9, respectively. The mean summed DISCERN score was 39.6 ± 2 out of 80. Claude-3-sonnet was considered of poor quality while others were fair. The mean accuracy score of the LLM-chatbots was 4.4 ± 0.4 out of 5. The three LLMs performed similarly on readability and accuracy. The LLM-chatbots generally scored well in providing treatment options and their mechanisms, while lacking references and information regarding treatment benefits, risks of alternative treatments and impact on quality of life.
Conclusions: The three LLM-chatbots showed comparable and satisfactory accuracy in providing information on pediatric phimosis; however, their readability and overall quality were suboptimal. These limitations may affect parent counseling, shared decision-making and informed consent. LLM-generated information should, therefore, supplement, not replace, clinician guidance and reliable medical resources.
Level of evidence: III.

Keywords:  Information Technology; Patient Participation

DOI:  https://doi.org/10.1136/wjps-2026-001276
J Orthop Surg (Hong Kong). 2026 Sep-Dec;34(3):34(3): 10225536261504909

Clinical accuracy, information quality and readability of chatbot responses to patient questions about lower extremity amputation.

Kamil Balaban, Mehmet Batu Ertan, Mahmut Kalem.

  ObjectiveTo compare the clinical accuracy, information quality and readability of chatbot responses to common patient questions about lower extremity amputation.MethodsTwenty frequently asked patient questions were submitted once to each of ChatGPT, Gemini and DeepSeek in separate sessions. Two reviewers, blinded to chatbot identity, independently assessed responses using a four-category clinical accuracy scale and the DISCERN instrument. Readability was quantified as the mean grade level across the Flesch-Kincaid Grade Level, Gunning Fog, Coleman-Liau, and Simple Measure of Gobbledygook indices. Comparisons used Friedman tests and Holm-adjusted Wilcoxon signed-rank tests.ResultsSixty responses were analysed. Mean (±standard deviation) clinical accuracy scores were 2.20 ± 0.57, 1.55 ± 0.65 and 1.20 ± 0.38 for ChatGPT, Gemini and DeepSeek, respectively (lower scores indicate greater accuracy). Gemini and DeepSeek received lower scores than ChatGPT (adjusted p=0.012 and p=0.002, respectively), whereas the Gemini-DeepSeek difference was not significant (p=0.090). Mean DISCERN scores were 50.05 ± 5.78, 53.33 ± 5.43 and 58.23 ± 3.39, respectively, with significant differences in all pairwise comparisons. Mean reading grades were 12.93 ± 1.28, 13.17 ± 1.36 and 11.74 ± 1.15, respectively; DeepSeek had lower grades than both comparators (both adjusted p<0.001).ConclusionDeepSeek responses received higher information quality scores and had lower reading grade levels, although clinical accuracy did not differ significantly between DeepSeek and Gemini. All three chatbots exceeded commonly recommended reading levels for patient education. Clinician review and language adaptation are warranted before chatbot-generated information is used to supplement individualised counselling on amputation and rehabilitation.

Keywords:  Amputation; Consumer Health Information; Large Language Models; Lower Extremity; Patient Education; Surgical

DOI:  https://doi.org/10.1177/10225536261504909
J Eval Clin Pract. 2026 Dec;pii: e70721. [Epub ahead of print]32(8):

Quality, Readability, and Reliability of Artificial Intelligence Chatbot Responses to Parent-Centered Questions on Newborn Hearing Screening: A Cross-Sectional Comparative Study.

Arif Ozses, Beyza Demirtas Yilmaz, Merve Ozses.

   AIMS AND OBJECTIVES: Parents increasingly use large language models (LLMs) to obtain health-related information; however, the quality, readability, and reliability of AI-generated patient education remain uncertain. This study evaluated the performance of three LLM-based chatbots in providing parent-centred information regarding newborn hearing screening.
METHODS: Thirty-eight parent-centred questions regarding newborn hearing screening were submitted to ChatGPT-5.1, Gemini 2.5 Flash, and DeepSeek-V3. Responses were assessed using a five-point Likert scale for accuracy, comprehensiveness, clarity, and credibility. Responses were assessed using the modified EQIP for quality, and the Flesch Reading Ease Score (FRES) and Flesch-Kincaid Grade Level (FKGL) for readability.
RESULTS: DeepSeek-V3 outperformed both ChatGPT-5.1 and Gemini 2.5 Flash in accuracy and comprehensiveness (p < 0.05), whereas its clarity scores were lower (p < 0.05). The FRES scores (51.5-59.3) indicated difficult readability, whereas the FKGL scores (8.4-8.9) corresponded to an eighth-grade reading level. Gemini 2.5 Flash demonstrated higher FRES scores (p = 0.001). mEQIP scores were not significant (p = 0.148) and indicated good quality.
CONCLUSIONS: LLM-based chatbots generally perform acceptably on newborn hearing screening questions, although substantial variability was observed across models. Despite acceptable readability levels, chatbot-generated content remains relatively difficult to read. LLMs may serve as complementary information resources for parents seeking information about newborn hearing screening, although further research is needed before their routine clinical use.
RATIONALE: This study addresses the JECP journal's aim to evaluate the clinical applications of emerging artificial intelligence technologies in healthcare. Evaluating the quality, accuracy, readability, and reliability of chatbot-generated information for newborn hearing screening provides evidence to support person-centred care, patient education, and the responsible integration of AI into clinical practice.

Keywords:  artificial intelligence; health literacy; large language models; newborn hearing screening; patient education

DOI:  https://doi.org/10.1111/jep.70721
Cleft Palate Craniofac J. 2026 Nov 05. 10556656261499122

Online Information Available to Caregivers on Introducing Solid Foods to Infants with Cleft Palate: A DISCERN-Based Cross-Sectional Analysis.

Evy I Loomis-Goltl, Lian Arzbecker, Jessica L Chee-Williams, Katelyn J Kotlarek.

  ObjectiveTo evaluate the availability, content, and quality of written online information regarding the introduction of solid food to infants with cleft palate.DesignCross-sectional analysis of publicly accessible internet sources.SettingGoogle search engine using standardized search terms.Main Outcome MeasuresContent characteristics including recommended age of solid food introduction, feeding strategies, and clinical guidance. Quality was assessed using the DISCERN instrument.ResultsOf 56 unique sources identified, 30 met inclusion criteria. Hospital/health system resources were the most common source type (40.0%), and four sources (13.3%) cited peer-reviewed literature. Recommendations for timing of solid food introduction varied, although most sources recommended introduction at approximately 6 months (60.0%). Most sources guided nasal regurgitation (83.3%), food textures and progression (83.3%), and feeding strategies or adaptations (56.7%). The mean DISCERN score was 58.5 (SD = 23.7; median = 71; range = 16-80), with substantial variability in quality across sources.ConclusionsOnline information regarding solid food introduction for infants with cleft palate is variable and often lacks evidence-based guidance. Improved dissemination of standardized recommendations is needed to support caregivers and optimize feeding outcomes in this population.

Keywords:  cleft palate; development; feeding

DOI:  https://doi.org/10.1177/10556656261499122
Cureus. 2026 Sep;18(9): e117217

Assessing the Readability and Quality of Online Patient Information Regarding Aspirin-Exacerbated Respiratory Disease.

Faiz A Shaikh, Huseyin Turacli, Rakesh Mistry, Raj Lakhani.

  Background The internet has revolutionised the way in which patients obtain information regarding their health. In order to be accessible, Health Education England recommend patient information should be at the reading age of an average 11-year-old. Effective management of aspirin-exacerbated respiratory disease (AERD) is complex and includes a combination of medical (including aspirin desensitisation and biological treatments) and surgical therapy to improve quality of life and reduce recurrence. We objectively assess the quality and readability of online patient materials on AERD. Methods The three search engines utilised were Google (Google LLC, Mountain View, CA, USA), Bing (Microsoft Corporation, Redmond, WA, USA) and Yahoo (Yahoo Inc., New York City, NY, USA). Three specific search terms ('Samter's Triad', 'ASA Triad', and 'AERD') were used to identify the top 40 websites from each search result. After applying eligibility criteria, a total of 30 websites were finally analysed. Selected websites were assessed for readability using the Flesch-Kincaid Reading Ease Score (FRES), Flesch-Kincaid Grade Level (FKGL), and the Gunning-Fog Index (GFI). The quality of information available online was also assessed using the DISCERN tool. Results The average FRES was 36.95, which correlates to a US grade level of 11.5 (16-17-year-olds). The average GFI score was 15.07, which equals that of a 21-year-old. According to the DISCERN criteria, the overall quality of the health information was poor. Conclusion Health information available on the internet regarding AERD is overall of poor quality. The information available is written above Health Education England's recommended reading level, which may reduce a patient's understanding of AERD. As more patients turn to the internet to obtain health information, it is of utmost importance that information is validated and evidence-based to counter the spread of misinformation.

Keywords:  aerd; aspirin-exacerbated respiratory disease; discern instrument; health literacy; online health information; patient education; readability; samter's triad

DOI:  https://doi.org/10.7759/cureus.117217
Front Oral Health. 2026 ;7 1975960

The use of different artificial intelligence-powered chatbots to answer questions related to dental implants.

Rasha A Alamoush, Mohammad Atrash, Ziad Abusidu, Alaa H A Sabrah, Islam Abd Alraheam, Marwa M Alnsour, Mahmoud K Al-Omiri.

   Background: Artificial intelligence (AI)-powered chatbots based on large language models (LLMs) are increasingly used as sources of health information. In dentistry, patients frequently seek online information regarding dental implant therapy; however, the quality, reliability, and clinical usefulness of AI-generated responses remain uncertain. This study aimed to compare the performance of commonly available AI chatbots in answering dental implant-related questions.
Methods: Ninety-nine publicly available dental implant-related questions were identified using online question-generation platforms and categorized into four domains: biological and surgical aspects, prosthetic and material science, clinical outcomes and complications, and patient-centered and economic considerations. Each question was submitted independently to four AI chatbots: ChatGPT-5, Gemini 3 Pro, DeepSeek V3.2, and Microsoft Copilot. Responses were evaluated by two calibrated dental specialists using the Global Quality Scale (GQS), CLEAR reliability assessment tool, usefulness scoring, and readability indices including Flesch-Kincaid Grade Level (FKG) and Simple Measure of Gobbledygook (SMOG). Statistical analyses were performed using appropriate parametric and non-parametric tests with a significance level of p < 0.05.
Results: All evaluated chatbots generated responses with very good reliability according to CLEAR scores. Significant differences were observed among AI models in readability, quality, usefulness, and reliability scores (p < 0.001). DeepSeek demonstrated the highest reliability (median CLEAR score = 25) and the highest proportion of excellent-quality responses according to GQS. ChatGPT generated significantly more useful responses compared with other models. DeepSeek and Copilot produced more readable responses, whereas Gemini demonstrated the highest reading difficulty based on FKG and SMOG scores. Response performance was not significantly influenced by the question category.
Conclusions: AI-powered chatbots can provide generally reliable information regarding dental implants and may serve as supportive sources of patient education. However, differences exist between AI models regarding readability, usefulness, and response quality. AI-generated information should complement rather than replace professional dental consultation, and further studies involving patients and clinically validated outcomes are warranted.

Keywords:  artificial intelligence; chatbots; dental implants; digital dentistry; health literacy; large language models; patient education

DOI:  https://doi.org/10.3389/froh.2026.1975960
Radiography (Lond). 2026 Nov 03. pii: S1078-8174(26)00295-6. [Epub ahead of print]32(8): 103619

Readability, understandability, actionability and information quality of online patient information on diagnostic imaging during pregnancy and breastfeeding: A cross-sectional study.

J Naidu, A Raguraman, S Naidu, A Prashani, H Ferreira, V Baskaradoss.

   INTRODUCTION: Patients commonly use online information when considering imaging during pregnancy or breastfeeding. Information that is difficult to read, poorly contextualised or inconsistent may contribute to anxiety, delayed investigation or unnecessary breastfeeding interruption. This study aimed to evaluate the readability, understandability, actionability and information quality of highly ranked online patient information on diagnostic imaging during pregnancy and breastfeeding.
METHODS: A cross-sectional Google search was performed on 30 July 2026 using 15 predefined patient-relevant search terms. The first 10 organic results for each term were screened. Duplicate occurrences were removed and 51 unique eligible resources were analysed at resource level. Readability was assessed using Flesch-Kincaid Grade Level (FKGL) as the primary metric, with Flesch Reading Ease (FRE), Simple Measure of Gobbledygook (SMOG), Coleman-Liau Index and Gunning Fog Index as supplementary measures. Understandability and actionability were assessed using the Patient Education Materials Assessment Tool for Printable Materials (PEMAT-P), and information quality using Journal of the American Medical Association (JAMA) benchmarks and DISCERN (quality of written consumer health information on treatment choices). Clinical content was reviewed descriptively using structured domains.
RESULTS: Mean FKGL was 8.6 (standard deviation [SD] 1.3); 49/51 resources (96.1%) exceeded the recommended sixth-grade level. Mean PEMAT-P understandability and actionability scores were 78.4% and 72.3%, respectively. Mean JAMA and DISCERN scores were 2.8/4 and 60.2. Inter-rater reliability was excellent for PEMAT-P understandability, PEMAT-P actionability and DISCERN [intraclass correlation coefficient (ICC) (2,1) = 0.982-0.998]. Structured narrative review identified variable risk contextualisation and inconsistencies in contrast-media and breastfeeding advice.
CONCLUSION: Highly ranked resources were generally understandable but frequently exceeded recommended reading levels, while transparency, risk contextualisation and practical clinical advice varied.
IMPLICATIONS FOR PRACTICE: Radiography services should review online patient information for reading demand, transparency, risk contextualisation and consistency of practical advice, with particular attention to maternal and foetal risk communication, contrast media and breastfeeding.

Keywords:  Breastfeeding; Diagnostic imaging; Health literacy; Patient education; Pregnancy; Readability

DOI:  https://doi.org/10.1016/j.radi.2026.103619
Health Info Libr J. 2026 Jun;pii: e70014. [Epub ahead of print]43(2):

Information on Greek Web Pages About Dietary Weight Loss Should Be More Accurate and Easier to Understand.

Evaggelia Fappa, Mary Micheli, Petroula Tsirpanli, Ioanna Pylarinou, Aristea Gioxari, George Ι Panoutsopoulos.

   BACKGROUND: Many people search for online dietary information, which seems to be of poor quality. Most studies have assessed such information published in English.
OBJECTIVES: To assess the quality and readability of online weight loss dietary information published in the Greek language.
METHODS: Nineteen queries retrieved 195 websites eligible for assessment for both quality and readability which were classified into commercial, news, blogs, personal/professional, forums, and portals. Overall information quality was assessed with the DISCERN tool (score: 16-80) and content accuracy with a checklist (score: 0-9) based on the latest EASO guidelines on weight loss. Readability was assessed by the Common European Framework of Reference (CEFR) level of language.
RESULTS: The median scores were 43/80 for the DISCERN instrument and 3/18 for the EASO guidelines checklist. More than 90% of websites were of at least B2 CEFR level. The quality of information varied for both DISCERN and EASO guidelines checklist scores between website categories.
DISCUSSION: Greek online weight loss information has fair overall quality, poor content accuracy, and is above the recommended readability level.
CONCLUSION: Greek online weight loss dietary information is not trustworthy and difficult to understand by lay audiences.

Keywords:  Greece; consumer health information; health literacy; information dissemination; public health; south‐east Europe; web sites

DOI:  https://doi.org/10.1111/hir.70014
J Craniofac Surg. 2026 Nov 04.

An Appraisal of Printed Online Education Materials on Transoral Robotic Surgery (TORS) Online Education for Patients.

Jed Speers, J W Spatafore, Milena Fabry, Shelby Kitchin, Sameep P Kadakia, Priti Parikh, Mohamedkazim Alwani.

   PURPOSE: As both patients and large language models (LLMs) utilize online health information about medical conditions and treatments, ensuring the quality of these resources is increasingly important. This study evaluated the readability, understandability, and quality of printed online education materials (POEMs) pertaining to transoral robotic surgery (TORS).
METHODS: Printed online education materials were identified using the Google term "transoral robotic surgery," with the first 50 websites selected for analysis. Readability was assessed using the Flesch-Kincaid Grade Level (FKGL), Flesch Reading Ease (FRE), and Gunning Fog Index (GFI). Understandability was evaluated using the Patient Education Materials Assessment Tool for Printable Materials (PEMAT-P), and overall quality was assessed using the DISCERN instrument.
RESULTS: Mean [SD] FKGL, FRE, and GFI scores were 11.35 [1.90], 41 [10], and 13.85 [2.30], respectively. The mean [SD] PEMAT-P understandability score was 67% [11%]. The median [interquartile range (IQR)] DISCERN quality score was 41 [36-48]. These findings indicate that POEMs about TORS frequently demonstrated limited readability, understandability, and quality.
CONCLUSION: Printed online educational materials related to TORS are often written at reading levels exceeding recommended standards and demonstrate suboptimal understandability and quality. Efforts to improve the readability, understandability, and quality of online TORS educational materials may enhance patient comprehension and support informed decision-making. Such improvements may also benefit the broader digital health information environment, as publicly available online resources increasingly serve as source material for AI-assisted health information tools.

Keywords:  Printed online education materials; TORS; readability; transoral robotic surgery; understandability

DOI:  https://doi.org/10.1097/SCS.0000000000013558
Brain Spine. 2026 ;6 107476

Quality and readability of online patient education materials for trigeminal Neuralgia: A cross-sectional review.

Philbert Yu Xuan Teo, Selnan Wuyep, Andreas K Demetriades.

   Background/objectives: Patients with trigeminal neuralgia frequently seek online information about symptoms, diagnosis and treatment options. Previous work has identified weaknesses in the quality of online trigeminal neuralgia information and high estimated reading demands. This study evaluated the quality and readability of online patient education materials on trigeminal neuralgia.
Methods: Google was searched for the term "trigeminal neuralgia," and the first 30 results were screened, assessing the linked webpage rather than the parent website. English-language, freely accessible, patient-facing webpages were included. Two reviewers independently assessed quality using the DISCERN instrument, and item-level scores were averaged for analysis. Readability was assessed using standard readability formulae from ReadabilityFormulas.com after removal of non-educational webpage content.
Results: Of 30 screened results, 22 webpages were included. Included webpages comprised 10 general health information sites and 12 healthcare provider sites. The mean Total DISCERN score across reviewers was 49.32 ± 9.43, and the mean overall DISCERN rating was 3.30 ± 0.70. The lowest mean item scores concerned the consequences of no treatment (1.61), effects on quality of life (1.77), and clarity of evidence sources (2.14). The mean average reading level consensus was 10.93 ± 1.49, with 21 webpages (95.5%) exceeding grade 8 by this measure and none at or below grade 6.
Conclusions: The sampled online patient education materials on trigeminal neuralgia were commonly written above recommended reading levels and varied in DISCERN-assessed information quality. Clinicians should be aware that patients may encounter resources that are difficult to read or incomplete. Clinical accuracy, patient comprehension, and effects on treatment decisions were not assessed.

Keywords:  DISCERN; Internet; Patient education; Readability; Trigeminal neuralgia

DOI:  https://doi.org/10.1016/j.bas.2026.107476
Rheumatol Int. 2026 Nov 05. pii: 305. [Epub ahead of print]46(11):

Clinical accuracy and applicability of YouTube videos on rheumatic diseases and pregnancy.

Tugba Ozudogru Celik.



Keywords:  Information dissemination; Patient education as topic; Pregnancy; Rheumatic diseases; Social media

DOI:  https://doi.org/10.1007/s00296-026-06343-6
J Autism Dev Disord. 2026 Nov 04.

Quality and Reliability of YouTube Videos as a Source of Information Regarding Early Signs of Autism Spectrum Disorder.

Kadriye Ozyazici, Sibel Aydin, Oguz Emre, Eren Can.

   BACKGROUND: YouTube is a widely used source of health information, but the quality of videos on early signs of autism spectrum disorder (ASD) remains unclear.
METHODS: A cross-sectional content analysis of 50 English-language YouTube videos on early signs of ASD was conducted. Video quality and reliability were assessed using the Global Quality Scale (GQS), modified DISCERN, and JAMA benchmark criteria. Source Videos were categorized by source (Expert-created, parent-created, and mixed-source).
RESULTS: Expert-created and mixed-source videos scored significantly higher than parent-created videos on the modified DISCERN and GQS scales (p < .001 for both). On the JAMA criteria, the overall group difference was smaller (p = .028); only the mixed-source group differed significantly from the parent-created group, while expert-created videos did not differ significantly from either group. Most videos (74%) covered all four early symptom domains. No significant correlation was found between quality scores and popularity metrics (views, likes). The three source types did not differ significantly in the number of early-sign domains covered (H(2) = 2.628, p = .269). Effect sizes were large for DISCERN (ε² = 0.424) and GQS (ε² = 0.511).
CONCLUSIONS: YouTube videos on early signs of ASD vary widely in quality. Parent-created content is significantly less reliable. Clinicians and content developers should consider that expert-created, parent-created, and mixed-source videos serve complementary informational functions rather than a single hierarchy of quality. Co-produced content that combines professional expertise with parental narrative may offer practical value beyond what either source provides alone.

Keywords:  Autism spectrum disorder; Early signs; Quality assessment; YouTube

DOI:  https://doi.org/10.1007/s10803-026-07601-z
Arthrosc Sports Med Rehabil. 2026 Oct 30. e70136

YouTube Videos on Arthrogenic Muscle Inhibition Show Low-to-Moderate Quality and Reliability.

Jacopo Conteduca, Damiano Longo, Fabrizio Piacquadio, Alberto Casto, Luigi Valentino, Andrea Bernetti, Giuseppe Rollo.

Purpose: To evaluate the quality and reliability of YouTube videos on arthrogenic muscle inhibition (AMI) using validated assessment instruments and to identify predictors of higher-quality content.
Methods: This cross-sectional study analyzed English-language YouTube videos identified on November 2, 2025, using the search terms "arthrogenic muscle inhibition" and "AMI." Video quality was assessed using the DISCERN, Journal of the American Medical Association, and Global Quality Score instruments. Assessments were performed independently by 2 experienced knee surgeons. Correlations between quality scores and video characteristics (views, likes, comments, time online, duration, source, and topic) were analyzed using nonparametric tests and multivariate regression with Bonferroni correction.
Results: A total of 45 videos were analyzed. Mean scores were 44.9 ± 20.5 for DISCERN, 1.47 ± 1.56 for Journal of the American Medical Association, and 2.93 ± 1.53 for Global Quality Score, indicating overall low-to-moderate quality. Video quality was strongly correlated across scoring systems (ρ > 0.86, P < .001). Videos addressing mechanisms and therapeutic interventions had the highest quality (mean DISCERN score: 62.7), whereas videos on patient experience and rehabilitation alone scored lowest (P < .001). Comment count correlated negatively with DISCERN scores (ρ = -0.37, P = .046). Video topic was the strongest independent predictor of video quality (P < .001).
Conclusions: YouTube videos on AMI demonstrate low-to-moderate quality and reliability, with substantial variation by topic and content type.
Clinical Relevance: Improving the quality of online AMI-related content may enhance patient education and promote accurate dissemination of rehabilitation information.

DOI: https://doi.org/10.1002/ars2.70136
BMC Pulm Med. 2026 Aug 08. pii: 531. [Epub ahead of print]26(1):

Quality and reliability of youtube videos on CPAP, BiPAP, and non-invasive ventilation: a cross-sectional content analysis.

Tugba Menekli, Turgut Dolanbay.

   OBJECTIVE: Correct use of non-invasive ventilation (NIV) and positive airway pressure devices (CPAP/BiPAP) is essential for treatment effectiveness and patient adherence. This study aimed to evaluate the reliability, quality, and patient education-related characteristics of YouTube videos about CPAP, BiPAP, and NIV devices.
MATERIALS AND METHODS: This cross-sectional content analysis evaluated YouTube videos related to CPAP, BiPAP, and NIV devices. A YouTube search was conducted between January 10 and February 10, 2026, using the keywords "CPAP tutorial," "BiPAP settings," and "non-invasive ventilation device." The first 50 results for each keyword were screened, and 73 videos met the inclusion criteria. Videos were categorized as individual or institutional uploads. Two independent experts evaluated the videos using DISCERN, the Journal of the American Medical Association (JAMA) benchmark criteria, the Global Quality Scale (GQS), and the Patient Education Materials Assessment Tool for Audiovisual Materials (PEMAT-A/V). Statistical analyses were performed using SPSS version 27.
RESULTS: Among the 73 videos, 49.3% were uploaded by individual users and 50.7% by institutional sources. According to PEMAT-A/V, 60.3% of the videos demonstrated low levels of understandability and actionability. Median scores were GQS 4 (3-4), DISCERN 57 (49-64), and PEMAT-A/V 68 (63-74). Institutional videos had significantly higher subscriber counts, views, and GQS, DISCERN, and PEMAT-A/V scores. Video duration showed positive correlations with views, likes, and DISCERN scores. Inter-rater reliability was high for DISCERN and PEMAT-A/V but poor for JAMA; therefore, JAMA-related findings should be interpreted cautiously.
CONCLUSION: YouTube videos related to CPAP, BiPAP, and NIV generally demonstrated moderate quality, but many provided limited patient education value. Content produced by institutional sources was associated with higher quality and better patient education metrics. These findings highlight the importance of developing more standardized and patient-centered online educational materials.
TRIAL REGISTRATION: Not applicable.

Keywords:  BiPAP; CPAP; Health information quality; Non-invasive ventilation; Patient education; YouTube

DOI:  https://doi.org/10.1186/s12890-026-04569-5
Front Digit Health. 2026 ;8 1964412

Quality and reliability of short videos about type A aortic dissection on Bilibili and TikTok: a cross-sectional study.

Keyu Wei, Mengqi Liu, Yuji Yan, Ruiying Lan, Yizi Zhang, Xiaolei Han, Ying Wang, Wenxu He, Xinhua Tang, Naizhong Zhang, Xiaolin Gao, Chunyu Tian, Xiaoling Wang.

   Background: Type A aortic dissection (TAAD) is a life-threatening cardiovascular emergency. Short-video platforms such as TikTok and Bilibili have become important channels for health information dissemination, but the quality and reliability of TAAD-related videos remain unclear.
Objective: This study aimed to evaluate the quality and reliability of TAAD-related short videos on TikTok and Bilibili, with particular attention to platform differences and uploader characteristics.
Methods: The top 250 videos ranked by comprehensive ranking on each platform were screened using the Chinese keyword "A." A total of 391 eligible videos were included. Uploaders were classified as medical practitioners, individual users, or institutions. Video quality and reliability were assessed using the Global Quality Score (GQS), modified DISCERN (mDISCERN), JAMA benchmarks, and Video Information Quality Instrument (VIQI). Multivariable regression analyses were performed to identify factors independently associated with video quality.
Results: TikTok videos showed higher engagement, whereas Bilibili videos were longer and provided broader coverage of epidemiology, etiology, and clinical presentation-related content. TikTok remained independently associated with higher VIQI scores after adjustment. Medical practitioners and institutions generally achieved higher VIQI and JAMA scores than individual users, and content coverage score was independently associated with higher video quality. However, mDISCERN scores did not differ significantly among uploader types.
Conclusion: TAAD-related short videos demonstrated moderate overall quality but limited reliability. Platform characteristics, uploader type, and content coverage were associated with assessed information quality. Professional participation may improve some aspects of health communication, but uploader identity alone does not ensure higher reliability or transparency.

Keywords:  Bilibili; GQS; TikTok; mDISCERN; type A aortic dissection; video quality

DOI:  https://doi.org/10.3389/fdgth.2026.1964412
J Robot Surg. 2026 Oct 31. pii: 1161. [Epub ahead of print]20(1):

Information quality, clinical content, and engagement of robot-assisted nephrectomy videos on Bilibili and Douyin: a cross-sectional study.

Kai Hui, Na Zhao, Wenjie Wan, Huajie Gao, Guangcheng Dai, Yaowen Wang.

  Surgical process information is increasingly being obtained over video platforms. Videos on robot-assisted nephrectomy are still ambiguous in terms of quality and dependability. To assess information quality, clinical completeness and accuracy, understandability, actionability, and engagement, while treating platform comparisons as exploratory. On September 10, 2026, this cross-sectional study screened the first 250 search results from each of Bilibili and Douyin using the same search phrase. Eligible videos were assessed with the Global Quality Scale (GQS), a study-specific five-item Likert adaptation of mDISCERN, and JAMA benchmark criteria. A ten-domain clinical checklist assessed completeness and standardized accuracy. The Patient Education Materials Assessment Tool for Audiovisual Materials (PEMAT-A/V) assessed understandability and actionability. We examined group differences and Spearman correlations with engagement. Of 500 screened results, 121 videos were included (27 Bilibili, 94 Douyin). Median GQS was 3.0 (IQR, 2.5 to 3.5), and median clinical completeness was 10.0 of 20 (IQR, 5.5 to 12.5). Median PEMAT-A/V understandability and actionability were 54.5% and 37.5%, respectively. Professional-source videos had higher median GQS (3.5 vs. 2.5) and clinical completeness (11.5 vs. 5.5) than nonprofessional-source videos (both adjusted P < 0.001). Among videos with evaluable domains, median checklist-based standardized accuracy was 80.0% versus 25.0% in these groups (79 vs. 35 videos; adjusted P < 0.001). Commercial-promotion videos had the lowest median GQS (2.0) and clinical completeness (2.5). Composite engagement correlated weakly with GQS and clinical completeness (Spearman rho = 0.09 for each). Commercial promotion accounted for 1/27 Bilibili videos and 22/94 Douyin videos. Engagement was a poor proxy for assessed quality. The unequal content mix prevents platform-level attribution of the observed differences. Clearer evidence sourcing, fuller clinical coverage, and more actionable guidance are needed.

Keywords:  Health information quality; JAMA benchmark; Modified DISCERN; Robot assisted nephrectomy; Short video; Social media

DOI:  https://doi.org/10.1007/s11701-026-04121-x
BMC Oral Health. 2026 Aug 08. pii: 2181. [Epub ahead of print]26(1):

Information quality of implant-related videos on Chinese social media platforms: a cross-sectional analysis of Bilibili, Douyin, and Xiaohongshu.

Shuhan Gu, Ziyue Xu, Qianqian Hou, Yiming Gao, Kaixiao Yan.

   BACKGROUND: As dental implant therapy becomes increasingly common, patients frequently seek implant-related information through social media platforms. However, the completeness and educational quality of such information remain uncertain, particularly on Chinese social media platforms. This study aimed to evaluate implant-related videos on three major Chinese social media platforms using complementary measures of content completeness and educational quality.
METHODS: A cross-sectional content analysis was conducted on implant-related videos retrieved from Bilibili, Douyin, and Xiaohongshu on June 10, 2026, using the Chinese keyword "" ("dental implant"). The first 100 videos returned by each platform's default search function were screened, and 250 videos meeting predefined eligibility criteria were included. Video characteristics were recorded, and information quality was evaluated using the Global Quality Scale (GQS) and a newly developed Implant Content Score (ICS), a 10-domain instrument assessing the completeness of implant-related information. Differences among platforms and associations between video characteristics and information quality were statistically analyzed.
RESULTS: Of the 250 included videos, 83 originated from Bilibili, 79 from Douyin, and 88 from Xiaohongshu. Significant differences in information quality were observed among platforms. Bilibili videos showed the highest median ICS score [4 (2-5)], followed by Xiaohongshu [3 (2-4)] and Douyin [2 (1-3)] (P < 0.001). Similarly, Bilibili demonstrated the highest GQS score, followed by Douyin and Xiaohongshu (P < 0.001). Scientific accuracy was the most frequently reported content domain (97.2%), whereas comprehensive treatment planning was the least frequently reported (8.8%). Video duration was positively associated with ICS score (r = 0.490, P < 0.001), while likes, comments, and favorites showed no significant association with information quality.
CONCLUSIONS: The quality of implant-related information on Chinese social media platforms varies substantially. Although Bilibili videos demonstrated relatively higher content completeness and educational quality, clinically important topics such as comprehensive treatment planning were frequently overlooked across all platforms. These findings suggest that content completeness and educational quality represent distinct dimensions of online health information quality and support the use of complementary assessment approaches. Greater involvement of dental professionals in developing evidence-based implant-related educational content may improve patient understanding and support informed treatment decision-making.

Keywords:  China; Dental implants; Health information; Patient education; Social media; Video analysis

DOI:  https://doi.org/10.1186/s12903-026-09555-1
Aesthetic Plast Surg. 2026 Nov 02.

Gynecomastia Trends on TikTok.

Nathaniel A Cleri, Averill Clapp, Evan Rothchild, Joseph A Ricci.

   BACKGROUND: Gynecomastia can cause significant distress, especially in young men. TikTok is a rapidly growing social media platform and information source for this demographic, but the platform's content and quality can be unreliable. This study assesses the quality and engagement of gynecomastia-related content on TikTok and evaluates the risk of misinformation using validated and widely used scoring tools.
METHODS: This study analyzed English-language gynecomastia-related TikTok videos. Search parameters included ten relevant hashtags, and the top twenty-five videos per hashtag (or maximum available) were analyzed. Video viewing metrics and quality assessment (validated mDISCERN and PEMAT tools) were collected and calculated. Videos were categorized by creator and video type.
RESULTS: The hashtags "#gynecomastiasurgeon" (84%) and "#gynecomastiasurgery" (64%) had the highest representation of videos by plastic surgeons. Patient videos had more comments compared to physician-created videos (p = 0.004). Lay/other-created videos scored lower on quality and understandability (mDISCERN, p < 0.001; PEMAT, p < 0.001). Content creator type was associated with mDISCERN scores (p < 0.001), while video type was associated with PEMAT scores (p < 0.001). Neither factor was associated with engagement score (p = 0.484). Overall, lay/other category videos had relatively high engagement, but lower-quality scores (mDISCERN and PEMAT), suggesting a greater risk of misinformation.
CONCLUSIONS: Patient-created videos garnered higher comment engagement, while lay/other videos demonstrated lower-quality content, highlighting the need for healthcare professionals to create high-quality content on TikTok to counter the risk of misinformation. When information is gathered and disseminated accurately, social media platforms can serve as wide-reaching methods for patient engagement and education.
LEVEL OF EVIDENCE V: This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors https://www.springer.com/00266 .

Keywords:  Gynecomastia; Misinformation; PEMAT; Social media; TikTok; mDISCERN

DOI:  https://doi.org/10.1007/s00266-026-06397-8
Eur J Ophthalmol. 2026 Nov 01. 11206721261504262

Quality and patient-safety content of Instagram videos on ophthalmic emergencies: A cross-sectional content analysis.

Fatih Ali Ceylan, Pınar Gidiş.

  PurposeTo evaluate the quality, reliability, and safety-related content of Instagram videos on common ophthalmic emergencies.MethodsThis cross-sectional content analysis included 175 publicly available English-language Instagram videos across eight ophthalmic emergency topics. Two ophthalmologists independently assessed videos using the Journal of the American Medical Association (JAMA) Benchmark Criteria, Global Quality Score (GQS), and a study-specific exploratory Eye Safety Score (ESS). Misinformation, harmful advice, emergency referral guidance, red-flag information, creator type, and engagement were recorded. Adjusted ordinal regression evaluated associations between engagement and ESS while accounting for topic, creator category, and repeated creators.ResultsMean JAMA, GQS, and ESS values were 2.55 ± 0.52, 2.96 ± 0.77, and 3.07 ± 1.18. Misinformation occurred in 10.9% of videos, harmful advice in 9.7%, emergency referral guidance in 64.0%, and red-flag information in 63.4%. Ophthalmologist-authored videos had higher reliability and quality scores and no detected misinformation in this sample. Views and likes showed weak inverse correlations with ESS; adjusted associations remained significant (views: adjusted odds ratio [aOR] = 0.47, 95% confidence interval [CI] 0.29-0.75; likes: aOR = 0.38, 95% CI 0.22-0.66).ConclusionIn this cross-sectional sample of English-language Instagram videos on ophthalmic emergencies, safety-related omissions were common. Higher engagement was associated with lower ESS, but the cross-sectional design precludes causal inference.

Keywords:  Instagram; digital health; emergency referral; health communication; misinformation; ophthalmic emergencies; patient safety; social media

DOI:  https://doi.org/10.1177/11206721261504262
Cureus. 2026 Oct;18(10): e117376

Social Media Literacy and Health Information-Seeking Behavior Among School Students and Educators in India: A Needs Assessment Under Project Info-Suraksha.

Rajni Kant, Bhavna B Mukhopadhyay, Nancepreet Kaur, Swapnil Jain, Saritha Nair, Enna Dogra Gupta, Shireen Fatma, Nhu Nguyen Tran Minh, Sophia Lonappan, Neha Srivastava.

  Purpose The increasing use of social media has made social media literacy vital for promoting safe, responsible, and informed online engagement. This study aimed to evaluate the knowledge, behaviors, and practices of school students and educators regarding social media usage and online health information seeking. The central research question focused on understanding how effectively students and educators navigate digital spaces and assess the credibility of online health information. Methods A cross-sectional mixed-methods needs assessment study was conducted across five Indian cities, encompassing 15 schools per city representing public, semi-private, and private institutions. The total sample included 2,331 students (grades 9-12) and 373 educators. A mixed-method approach, integrating quantitative surveys and qualitative interviews, was used. Results Internet access was reported by 97.5% of students and all educators. Among students, WhatsApp (Meta Platforms, Inc., Menlo Park, California, United States) (88.6%), YouTube (Alphabet Inc., Mountain View, California, United States) (85.2%), and Instagram (Meta Platforms, Inc.) (64%) were the most used platforms. Both social media (60.5%) and traditional media (58.5%) were primary sources of health information. While 88.4% of educators and 77.7% of students searched for health news online mainly via WhatsApp, YouTube, Instagram, and Facebook (Meta Platforms, Inc.), 33.2% of educators were unsure about information accuracy, but 55% still shared it. Only 4-6% of participants had received prior training on identifying misinformation. Conclusion The findings underscore the urgent need for structured capacity-building programs that enhance social media literacy, promote critical evaluation of online health content, and foster digital citizenship to mitigate misinformation and encourage responsible online behavior.

Keywords:  disinformation; education intervention; fake news; infodemic; misinformation; need assessment; online health information; social media literacy; training of teachers

DOI:  https://doi.org/10.7759/cureus.117376
J Med Internet Res. 2026 Oct 27. 28 e97878

Evaluating the Performance of Five Leading Large Language Models in Generating Health Education Materials for Patients With Infertility: Comparative Analysis.

Renjie Lu, Yuchen Wei, Siyuan Wu, Yaning Han, Shenyu Zhao, Lingmin Hu, Yang Chen.

   Background: Patients with infertility often have inadequate reproductive health literacy (RHL), while existing health education materials may be difficult to read, overly complex, and insufficiently tailored to individual needs. In recent years, large language models (LLMs) have created new opportunities to improve health education through their ability to generate and interpret medical information. However, their applicability and content quality in infertility education remain insufficiently evaluated.
Objective: This study systematically evaluated the readability and content quality of LLM-generated health education materials for patients with infertility, identified model strengths and limitations, and provided evidence to support their standardized use in reproductive health.
Methods: This study selected DeepSeek-R1, ChatGPT (GPT-5.2 Plus), Kimi 2.5, Qwen 3, and Gemini 3 as the study subjects. Based on the theoretical framework of RHL, this study developed a standardized question list comprising 6 dimensions and 93 items, and used unified prompts to generate health education texts in both Chinese and English. In this study, the Datayze Readability Analyzer was used to assess readability via the Flesch Reading Ease, Gunning Fog Index, and Flesch-Kincaid Grade Level (FKGL). Meanwhile, the content quality was evaluated in a double-blinded manner by reproductive medicine experts across 3 dimensions: accuracy, comprehensiveness, and humanistic care. Repeated-measures analysis of variance was used to compare performance across the LLMs, and Pearson correlation analysis was conducted to examine the associations between readability and content quality. A radar chart was used to evaluate the overall performance of each model.
Results: The 5 LLMs differed significantly in both text readability and content quality. In terms of readability, Gemini 3 and Qwen 3 performed best, with significantly higher FRE scores than the other 3 models. DeepSeek-R1 and Kimi 2.5 performed least favorably, with FKGL scores corresponding to a ninth-grade reading level. ChatGPT (GPT-5.2 Plus) showed intermediate performance. In terms of content quality, Kimi 2.5 performed best across accuracy, comprehensiveness, and humanistic care, followed by DeepSeek-R1. ChatGPT (GPT-5.2 Plus) performed least favorably. Analysis revealed significant negative correlations between readability and content comprehensiveness for texts generated by Kimi 2.5, indicating a trade-off between these 2 attributes. The composite radar chart analysis showed that Kimi 2.5 achieved the highest score (11,594.33), whereas ChatGPT (GPT-5.2 Plus) had the lowest score (5555.83).
Conclusions: Different LLMs showed significant differences in performance in generating health education texts for patients with infertility. In some models, readability was negatively correlated with content comprehensiveness. Under the conditions of this study, Kimi 2.5 demonstrated strong overall performance. However, no model achieved the best performance across all evaluation dimensions. Future applications of LLMs in reproductive health education will require domain-specific optimization, context-specific adaptation, and systematic quality control to support more accurate, comprehensive, and patient-centered communication.

Keywords:  content quality; health education; infertility; large language models; readability

DOI:  https://doi.org/10.2196/97878