Objective: To develop QingNangTCM, a specialized large language model (LLM) tailored for expert-level traditional Chinese medicine (TCM) question-answering and clinical reasoning, addressing the scarcity of domain-specific corpora and specialized alignment. Methods: We constructed QnTCM_Dataset, a corpus of 100 000 entries, by integrating data from ShenNong_TCM_Dataset and SymMap v2.0, and synthesizing additional samples via retrieval-augmented generation (RAG) and persona-driven generation. The dataset comprehensively covers diagnostic inquiries, prescriptions, and herbal knowledge. Utilizing P-Tuning v2, we fine-tuned the GLM-4-9B-Chat backbone to develop QingNangTCM. A multi-dimensional evaluation framework, assessing accuracy, coverage, consistency, safety, professionalism, and fluency, was established using metrics such as bilingual evaluation understudy (BLEU), recall-oriented understudy for gisting evaluation (ROUGE), metric for evaluation of translation with explicit ordering (METEOR), and LLM-as-a-Judge with expert review. Qualitative analysis was conducted across four simulated clinical scenarios: symptom analysis, disease treatment, herb inquiry, and failure cases. Baseline models included GLM-4-9B-Chat, DeepSeek-V2, HuatuoGPT-II (7B), and GLM-4-9B-Chat (freeze-tuning). Results: QingNangTCM achieved the highest scores in BLEU-1/2/3/4 (0.425/0.298/0.137/0.064), ROUGE-1/2 (0.368/0.157), and METEOR (0.218), demonstrating a balanced and superior normalized performance profile of 0.900 across the dimensions of accuracy, coverage, and consistency. Although its ROUGE-L score (0.299) was lower than that of HuatuoGPT-II (7B) (0.351), it significantly outperformed domain-specific models in expert-validated win rates for professionalism (86%) and safety (73%). Qualitative analysis confirmed that the model strictly adheres to the “symptom-syndrome-pathogenesis-treatment” reasoning chain, though occasional misclassifications and hallucinations persisted when dealing with rare medicinal materials and uncommon syndromes. Conclusion Combining domain-specific corpus construction with parameter-efficient prompt tuning enhances the reasoning behavior and domain adaptation of LLMs for TCM-related tasks. This work provides a technical framework for the digital organization and intelligent utilization of TCM knowledge, with potential value for supporting diagnostic reasoning and medical education.

Objective: To investigate the distribution frequency and molecular mechanism of the rare blood type Lu(a-b-) in Shenzhen, and to compare the polymorphisms of the Lutheran blood group system encoding gene LU and the In (Lu) phenotype-related gene KLF1 among Han Chinese, Indian, and Uyghur populations in Xinjiang. Methods: Serological methods were used to screen the Lu(a-b-) phenotype of blood donors in Shenzhen. Third-generation sequencing was employed to sequence the full-length of the LU and KLF1 genes in Lu (a-b-) phenotype samples as well as the samples from the Han Chinese, Indians, and Uyghur population, followed by analysis of gene haplotypes frequencies. Results: Ten individuals with the Lu(a-b-) phenotype were screened out of 14 367 blood donors in Shenzhen, yielding a frequency of approximately 0.07%. Only 2 cases showed mutations in the coding region of the LU gene, while all individuals showed heterozygous mutations in the coding region of the KLF1 gene. The highest mutation frequencies of the LU and KLF1 genes were observed in the Uyghur population in Xinjiang and the Han Chinese in Shenzhen, respectively. Conclusion: All Lu(a-b-) phenotypes are of the In (Lu) type, and their formation mechanism is mainly related to KLF1 gene mutations. Both the LU and KLF1 genes exhibit significant polymorphism in the Han Chinese, Indians, and Uyghur populations.

ObjectiveTo evaluate the efficacy and safety of Janus kinase （JAK） inhibitors combined with Chinese herbal medicine （CHM） in treating rheumatoid arthritis （RA）. MethodsClinical data from 169 RA patients were retrospectively collected. Among them， 71 cases received JAK inhibitors as the control group， while 98 cases received JAK inhibitors plus CHM as the observation group， both treated for 24 weeks. The rheumatoid factor （RF）， cyclic citic peptide antibody （anti-CCP）， erythrocyte sedimentation rate （ESR）， C-reactive protein （CRP）， and white blood cell count （WBC） were recorded before and after treatment. Databases including CNKI， Wanfang， VIP， PubMed and Web of Science were searched from inception till August 31st， 2025 for randomized controlled trials （RCTs） on the combined use of JAK inhibitors and CHM for RA. The methodological quality of the included studies was evaluated using the risk of bias assessment tool. Meta-analyses were performed for RF， anti-CCP， ESR， CRP， 28-joint disease activity score （DAS28）， overall clinical effective rate， and incidence of adverse events. Sensitivity analysis were also performed. ResultsThe retrospective study demonstrated that after treatment， ESR， CRP， and anti-CCP levels decreased in the observation group， while ESR and CRP levels decreased in the control group （P＜0.05）. Moreover， ESR and RF levels in the observation group were lower than those in the control group （P＜0.05）. A total of 9 RCTs involving 770 patients were included in the meta-analysis. The results indicated that the JAK inhibitors plus CHM group was superior to the JAK inhibitors group in reducing RF （MD=-8.97， 95%CI -15.01 to -2.94， P=0.004）， CRP （MD=-3.34， 95%CI -3.82 to -2.86， P＜0.001）， ESR （MD=-5.33， 95%CI -7.98 to -2.69， P＜0.001）， and DAS28 score （MD=-0.54， 95%CI -0.74 to -0.34， P＜0.001）， as well as in improving the overall clinical effective rate （OR=4.53， 95%CI 2.55 to 8.03， P＜0.001）. No statistically significant differences were observed between groups in anti-CCP levels （SMD=-2.08， 95%CI -4.41 to 0.24， P=0.080） or incidence of adverse events （OR=0.93， 95%CI 0.55 to 1.57， P=0.790）. ConclusionThe combination of JAK inhibitors and CHM demonstrates remarkable efficacy in treating RA， contributing to improved disease activity and reduced inflammatory markers with a favorable safety profile.

OBJECTIVE To construct a research question system of the Guideline for the Management of Therapeutic Drug Monitoring （hereinafter referred to as the Guideline ）， so as to provide the basis for the formulation of the Guideline . METHODS Based on literature research and expert interviews， the questionnaire about research questions of the Guideline was initially constructed. Through the modified Delphi method， the online questionnaire survey and expert consensus meeting were conducted among the Guideline development experts to determine the research question system of the Guideline . RESULTS A total of 59 questionnaires were distributed， and 52 valid questionnaires were collected. The positive coefficient of experts was 88.14%. The experts came from 21 provinces/autonomous regions/municipalities directly under the central government， covering multidisciplinary experts related to therapeutic drug monitoring， all of whom held senior professional titles. After one round of survey， a consensus was reached on the research questions， and a research question framework encompassing four research parts was constructed， including： indications for conducting therapeutic drug monitoring （4 first-level research questions，10 second-level research questions， 31 third-level research questions）； therapeutic drug monitoring technical process （3 first-level research questions， 9 second-level research questions， 13 third-level research questions）； result interpretation and clinical application （3 first-level research questions and 3 second-level research questions）； quality control （8 first-level research questions and 12 second-level research questions）. The importance scores for the four parts ranged from 4.71 to 4.88， with full-score rates all no less than 73.08%. The expert authori ty coefficients were all no less than 0.90， the coefficients of variation of importance scores were all no higher than 0.11， and Kendall’s W ranged from 0.464 to 0.626 （all P ＜0.05）. CONCLUSIONS The constructed research question system has high authority， scientificity， and reliability， laying a foundation for the standardized formulation of the Guideline .

The ethical review exemption system was introduced to meet the needs of modernized ethical governance， implement hierarchical management for ethical review projects involving human subjects， and explore an ethical management model that reduces procedural supervision for specific types of research involving no greater than minimal risk. Through a comparative study， this paper seeks to elaborate on the differences between China and the United States （US） in the ethical review exemption system from multiple aspects， such as ethical regulations， risk definition， preconditions， applicable scenarios， and operational procedures. The aim is to provide references for ethical review committees to better implement the ethical review exemption system， enhance policy research， and ensure the fairness and objectivity of ethical review. Through the comparative study of the ethics review exemption system in China and the US， the following recommendations are proposed for institutions implementing the ethical review exemption system for reference. First， the system should clearly define the meaning of ethical review exemption to prevent researchers or even committee members from misunderstanding it as falling outside the scope of ethical review， thereby exposing research participants’ rights and interests to unnecessary risks. Second， the system should specifically explain the preconditions， including the requirements for research types and research risks. Among them， risks should be considered from multiple dimensions such as physiology， psychology， society， and economy. Clear explanations should be provided for involving commercial interests and sensitive personal information. Finally， the system should establish basic principles and judgment criteria for applicable scenarios， including specifying exactly what the public data refers to and legal requirements that need to be met， providing a concept of anonymization to distinguish it from deidentified information data， and clarifying the relevant regulations and ethical principles to be followed， as well as the clear requirements for informed consent.

OBJECTIVE To further standardize the technical operations and management processes throughout therapeutic drug monitoring （TDM）， clarify the clinical value of TDM implementation， improve the scientific validity and reliability of monitoring results， and provide a solid reference basis for the formulation and optimization of clinical individualized precision dosing regimens. METHODS The Guidelines for the Management of Therapeutic Drug Monitoring were formulated in accordance with the latest definition of guidelines by the Institute of Medicine of the National Academies and the standard guideline development methodology of the World Health Organization， and in compliance with the requirements of the appraisal of guidelines for research and evaluation. A modified Delphi method was adopted to establish the research question system； evidence-based medicine research methods were applied to systematically search multiple databases to screen the latest and most comprehensive evidence. Evidence was graded and evaluated based on the evidence grading system of the Chinese Evidence-Based Medicine Center， and the grading criteria for recommendation strength from the Oxford Centre for Evidence-Based Medicine were used to determine the recommendation strength. The recommendation opinions were formed through multidisciplinary expert consensus. RESULTS The Guidelines for the Management of Therapeutic Drug Monitoring cover four core modules， including TDM application indications， technical procedures， result interpretation and clinical application， and quality control， involving 18 primary research questions， 34 secondary research questions， and yield 82 recommendations. CONCLUSIONS The guidelines systematically standardize the key technical links and management requirements of the whole TDM process， provide scientific and operable standardized tools， help improve the standardization level of TDM work， promote the translation of monitoring results into clinical decision-making， and provide strong support for precision personalized medicine and ensuring the safety and rationality of medication use.

This paper summarizes Professor SHI Zaixiang's clinical experience in treating high fever caused by sepsis using Chaihu Guizhi Ganjiang Decoction （柴胡桂枝干姜汤）. He holds that the key pathogenesis of sepsis involves constrained heat in the shaoyang and internal accumulation of water and fluids. The clinical manifestations such as high fever， chills， and alternating sensations of cold and heat are attributed to pathogenic heat constrained in the shaoyang. Meanwhile， soft tissue edema and serous cavity effusions are due to shaoyang dysfunction and internal water retention. In clinical practice， treating sepsis-related high fever requires addressing both the shaoyang-constrained heat and the associated edema and effusions. The therapeutic approach focuses on harmonizing the shaoyang and resolving internal fluids， using Chaihu Guizhi Ganjiang Decoction as the base formula with flexible modifications. Professor SHI emphasizes that this formula shows a rapid antipyretic effect， particularly in cases where multiple anti-infective treatments have failed.

OBJECTIVE To standardize the main processes and related technical links of the clinical comprehensive evaluation of drugs， and provide guidance and reference for improving the quality of comprehensive evaluation evidence and its transformation and application value. METHODS The construction of Guideline for the Workflow of Clinical Comprehensive Evaluation of Drugs was based on the standard guideline formulation method of the World Health Organization （WHO）， strictly followed the latest definition of guidelines by the Institute of Medicine of the National Academy of Sciences of the United States， and conformed to the six major areas of the Guideline Research and Evaluation Tool Ⅱ. Delphi method was adopted to construct the research questions； research evidence was established by applying the research methods of evidence-based medicine. The evidence quality classification system of the Chinese Evidence-Based Medicine Center was adopted for evidence classification and evaluation. The recommendation strength was determined by the recommendation strength classification standard formulated by the Oxford University Evidence-Based Medicine Center， and the recommendation opinions were formed through the expert consensus method. RESULTS & CONCLUSIONS The Guideline for the Workflow of Clinical Comprehensive Evaluation of Drugs covers 4 major categories of research questions， including topic selection， evaluation implementation， evidence evaluation， and application and transformation of results. The formulation of this guideline has standardized the technical links of the entire process of clinical comprehensive evaluation of drugs， which can effectively guide the high-quality and high-efficient development of this work， enhance the standardized output and transformation application value of evaluation evidence， and provide high-quality evidence support for the scientific decision-making of health and the rationalization of clinical medication.

OBJECTIVE To investigate the clinical characteristics of drug-induced liver injury （DILI） induced by glucagon- like peptide-1 receptor agonists （GLP-1RAs）， and to provide a reference for safe clinical medication. METHODS Using search terms such as “GLP-1”“GLP-1RAs”“semaglutide” “drug-induced liver injury”， relevant studies from PubMed， Embase， the Cochrane Library， CNKI， Wanfang Data and VIP were retrieved. Descriptive analysis was performed on cases of DILI induced by GLP-1RAs. RESULTS A total of 11 studies， comprising 11 patients， were included. Among them， 4 were male （36.4%） and 7 were female （63.6%）. Patient ages ranged from 17 to 64 years； 5 patients （45.5%） were between 50 and 65 years old. Six patients were treated for diabetes， and five for weight loss. Ten patients had underlying diseases. The shortest time to the onset of DILI was 5 days after medication， while the longest was approximately 180 days. The DILIs induced by GLP-1RAs were mainly hepatocellular injury type （6 cases）； severity levels included severe （3 cases）， moderate （6 cases）， and mild （2 cases）. Gastrointestinal symptoms and jaundice were the most common clinical manifestations. The association between DILI and GLP- 1RAs was assessed as “probable” in 10 cases and “possible” in 1 case. All 11 patients improved after drug discontinuation and （or） corresponding treatment. CONCLUSIONS DILI induced by GLP-1RAs is relatively concentrated in patients aged 50-65， with a higher incidence in females. The risk may be further increased in patients with underlying diseases. Clinical use of these agents should enhance pharmaceutical care， including identification of high-risk populations and patient education （especially symptom recognition）. When relevant symptoms appear， the drug should be discontinued immediately， with liver-protective therapy initiated when necessary， to ensure patient safety of drug use.

ObjectiveTo explore the effectiveness of feedforward control mode on improving occupational burnout among operating room nurses through theoretical research and clinical practice， with a view to promoting their physical and mental health and enhancing the quality of surgical nursing. MethodsA total of 440 operating room nurses from different regions， scales， and nursing experiences in Shaanxi Province from November 2023 to December 2023 were randomly divided into an experimental group and a control group， with 220 nurses in each group. While the control group received routine intervention measures， the experimental group introduced a feedforward control mode based on the control group， with “emotional exhaustion，” “depersonalization，”and“personal achievement” as observation indicators. ResultsThe incidence of occupational burnout in the experimental group was 11.4%， while that in the control group was 20.0%. The experimental group showed a significantly lower incidence than the control group （P=0.013）. ConclusionThe feedforward control mode can significantly alleviate nurses’ sense of job burnout， promote the improvement of surgical nursing quality， as well as continuously improve the scientific rigor， advancement， and humanistic nature of nursing services， which is conducive to building a harmonious and efficient nursing team. The spirit of teamwork， reflected in mutual support， mutual trust， and joint efforts for surgical success and patient health， has become an important component of nurses’ professional ethics.