Objective:To describe the hotspots and application trends of artificial intelligence (AI) in epidemiology in the past decade and analyze its advantages and challenges.Methods:The literatures with AI and epidemiology related keywords were systematically retrieved from Web of Science and China National Knowledge Infrastructure from 2014 to 2024. CiteSpace was used for bibliometric analysis of publication volume, keyword co-occurrence, clustering, emergence and cited literature co-occurrence analysis.Results:A total of 5 389 English papers and 1 659 Chinese papers were included, showing an increasing publication trend. High-frequency Chinese keywords included prediction, influencing factor, and machine learning, while English keywords frequently used were machine learning, prediction, and artificial intelligence. The Chinese keywords formed 14 clusters such as epidemiological characteristic, dietary pattern, and elderly individual, and the English keywords formed 21 clusters including prediction model, risk factor, and adult. In international studies, health policy, COVID-19, and digital health were the emerging frontier keywords. Eleven core papers were selected, covering key areas like traffic accident risk assessment, public health big data application, and deep learning in medical diagnosis.Conclusions:This study systematically summarized the research hotspots and development trends of AI applications in epidemiology over the past decade by using bibliometric methods, which indicated that current AI-based epidemiological studies are still in the exploratory phase, with the coexisting of both advantages and challenges. Continued attention should be paid to the future development of this field.

Objective To explore the current situation of medication safety behavior of elderly cancer patients and the path relationship of various influencing factors for improving medication safety behavior.Methods A total of 340 elderly cancer patients were investigated by a demographic questionnaire,the Medication Safety Behavior Scale,the Medication Literacy Scale,the Family Care Index Questionnaire,and the Chinese version of the Empowerment Scale for Cancer Patients from August to December 2024.The multiple linear regression analysis was applied to analyze influencing factors,and data were analyzed using SmartPLS 4.0 to construct a partial least squares structural equation model with path analysis.Results A total of 307 valid questionnaires were collected.The mean medication safety behavior score was 31.89±5.38.Residential area,drug literacy,family care,and health empowerment are factors that affect medication safety in elderly cancer patients,accounting for 37.3％of the total variation.The path analysis results indicated that health empowerment(β=0.480),medication literacy(β=0.154),and family care(β=0.227)positively correlate with medication safety behavior.Health empowerment played a partial mediating role between family care and medication safety behavior,as well as between medication literacy and medication safety behavior.The mediating effects are 0.125 and 0.332(P＜0.001),accounting for 35.51％and 68.31％of the total effect,respectively.Conclusion Medication safety behaviors among elderly cancer patients are at a median level and influenced by multiple factors.By improving their levels of health empowerment,healthcare professionals can motivate patients to take an active role in medication safety management.Further,promoting education on medication knowledge and teaching relevant medical skills,and together with guiding patients to perceive family care and support,can collectively improve their overall medication safety behaviors.

The emergence of artificial intelligence(AI)has had a significant impact on medical research and practice,both in terms of the number of studies and research paradigms,and has become an important tool for the development of pharmacoepidemiology.However,traditional AI has faced many challenges,while facilitating pharmacoepidemiology research,such as complex data processing,difficulty in identifying drug exposures and potential outcomes,and time-consuming and laborious study design and implementation.The rapid development of generative AI,represented by large language models(LLMs),has demonstrated a unique potential to enhance research efficiency,shift research paradigms,and facilitate knowledge discovery.LLMs are equipped with natural language understanding and generation capabilities.Through deep mining of multi-dimensional data resources,LLMs can quickly and accurately extract,analyze,summarize,and present the required information,which can not only help drug discovery,drug repurposing,pharmacovigilance and other pharmacoepidemiological tasks,but also provide powerful support for the whole process of research protocol design,data analysis,result interpretation and paper publication.Driven by LLMs,pharmacoepidemiology research is gradually moving into a new stage based on big data and automated analysis.Of course,LLMs also have problems of data bias,"illusion"of results,and ethical and legal regulation.By strengthening interdisciplinary cooperation,establishing a standardized evaluation system,improving ethical and regulatory guidance,enhancing data quality,strengthening practitioner training and capacity building,and promoting human-machine collaborative research modes,it is expected that the potential of LLMs in pharmacoepidemiology will be fully released,and it will provide a more scientific,rapid,and efficient technological support for drug regulation and public health decision-making.

Objective To investigate the association between the variability of remnant cholesterol inflammatory index(RCII),a novel composite biomarker,and the risk of stroke,in order to provide a theoretical basis for stroke prevention.Methods A prospective cohort study was conducted on 38 659 Kailuan individuals who took annual physical examinations in 2006,2008,and 2010.These subjects were grouped based on the quartiles of RCII variability,which was represented by standard deviation(SD)and average real variability(ARV),and were followed up every 2 years,with the occurrence of stroke(including ischemic and hemorrhagic strokes),death,or the end of follow-up on December 31,2022 as the endpoints.Kaplan-Meier method was used to calculate the cumulative incidence rate of endpoint events across different groups,and log-rank test was used to compare the difference of cumulative incidence of endpoint events in each group.Multivariate Cox proportional hazards regression model was adopted to analyze the association between RCII variability and risk of stroke.Results Among the 38 659 participants,a total of 2 539 strokes occurred during a mean follow-up period of 11.22±2.26 years.After adjusting confounding factors,when the participants were grouped by the quartiles of RCII-SD,the hazard ratio(HR)for stroke was 1.034(95%CI:0.917～1.167,P=0.584),1.146(95%CI:1.018～1.290,P=0.025),and 1.209(95%CI:1.066～1.370,P=0.003),respectively in the Q2,Q3,and Q4 groups,when compared with the Q1 group(Ptrend<0.05).When they were grouped by the quartiles of RCII-ARV,the HR for stroke was 1.008(95%CI:0.894～1.136,P=0.901),1.109(95%CI:0.986～1.248,P=0.085),and 1.152(95%CI:1.018～1.303,P=0.025),respectively,in the Q2,Q3,and Q4 groups,when compared with the Q1 group.Furthermore,both sensitivity and stratified analyses yielded similar results.Conclusion RCII variability is significantly associated with stroke,and the risk of stroke is gradually increasing with increment of the variability.Countermeasures Relevant authorities can focus on reducing RCII variability as a central objective by establishing regular monitoring mechanism,strengthening lifestyle interventions,and standardizing dietary,exercise,and weight management in order to suppress the index fluctuations.The principle of stable lipid-lowering in medication and optimization of therapeutic regimens with stable efficacy should be emphasized to prevent the risk of additional vascular damage.

Objective To explore the current situation of caring behavior among hospice nurses,and to analyze its latent profiles and population characteristics,so as to provide ideas for targeted interventions.Methods From August to November 2023,convenience sampling was used to select hospice nurses from 22 secondary and above hospitals in Jiangsu Province,Zhejiang Province,Shanghai City,Shandong Province,Anhui Province,Beijing City,Guangdong Province,and Sichuan Province.The demographic characteristics questionnaire,the Caring Behaviors Inventory,the Empathy Ability Scale for Hospice Nurses,and the Practice Environment Scale were used for investigation.Latent profile analysis was conducted based on 24 items of the Caring Behaviors Inventory as explicit indicators,and the influencing factors of different profiles were analyzed through multivariate logistic regression model.Results A total of 420 questionnaires were collected,of which 393 were valid,with a valid questionnaire response rate of 93.57％.The caring behavior of hospice nurses could be divided into 3 latent profiles,namely high level of care-low respect and connection group(49.62％),high to low caring behavior-overall fluctuation group(30.79％),and medium level of care-high knowledge and skills group(19.59％).The results of multivariate logistic regression showed that age,the dimensions of cognitive and emotional empathy in the Empathy Ability Scale for Hospice Nurses and the dimension of nursing foundations for quality of care in the Practice Environment Scale were the influencing factors of the latent profile of hospice nurses'caring behavior(P＜0.05).Conclusion There is significant heterogeneity in the caring behavior of hospice nurses.Nursing managers should develop individualized interventions for hospice nurses according to the influencing factors of different latent profiles to improve their level of caring behavior.

Grounded theory, as a flexible and systematic research method, serves as an important tool for gaining an in-depth understanding of clinical phenomena and nursing practice. This paper reviews the origin and development of grounded theory, its concepts and classifications, methodological procedures, and the necessity, significance, and current status of its application in the field of hospice care. The aim is to enhance the scientific application of grounded theory in hospice care research in China.

Grounded theory, as a flexible and systematic research method, serves as an important tool for gaining an in-depth understanding of clinical phenomena and nursing practice. This paper reviews the origin and development of grounded theory, its concepts and classifications, methodological procedures, and the necessity, significance, and current status of its application in the field of hospice care. The aim is to enhance the scientific application of grounded theory in hospice care research in China.

Decision making of artificial nutrition and hydration (ANH) for hospice care patients has been recognized as a complex and controversial issue that significantly impacted end-stage comfort and quality of life. This article reviewed the significance, status and influencing factors of decision making of ANH for hospice care patients. By analyzing the shortcomings of existing researches and clinical practices, it put forward the prospects for future research, so as to improve the decision-making dilemmas faced by hospice care patients.

The emergence of artificial intelligence(AI)has had a significant impact on medical research and practice,both in terms of the number of studies and research paradigms,and has become an important tool for the development of pharmacoepidemiology.However,traditional AI has faced many challenges,while facilitating pharmacoepidemiology research,such as complex data processing,difficulty in identifying drug exposures and potential outcomes,and time-consuming and laborious study design and implementation.The rapid development of generative AI,represented by large language models(LLMs),has demonstrated a unique potential to enhance research efficiency,shift research paradigms,and facilitate knowledge discovery.LLMs are equipped with natural language understanding and generation capabilities.Through deep mining of multi-dimensional data resources,LLMs can quickly and accurately extract,analyze,summarize,and present the required information,which can not only help drug discovery,drug repurposing,pharmacovigilance and other pharmacoepidemiological tasks,but also provide powerful support for the whole process of research protocol design,data analysis,result interpretation and paper publication.Driven by LLMs,pharmacoepidemiology research is gradually moving into a new stage based on big data and automated analysis.Of course,LLMs also have problems of data bias,"illusion"of results,and ethical and legal regulation.By strengthening interdisciplinary cooperation,establishing a standardized evaluation system,improving ethical and regulatory guidance,enhancing data quality,strengthening practitioner training and capacity building,and promoting human-machine collaborative research modes,it is expected that the potential of LLMs in pharmacoepidemiology will be fully released,and it will provide a more scientific,rapid,and efficient technological support for drug regulation and public health decision-making.

Objective:To describe the hotspots and application trends of artificial intelligence (AI) in epidemiology in the past decade and analyze its advantages and challenges.Methods:The literatures with AI and epidemiology related keywords were systematically retrieved from Web of Science and China National Knowledge Infrastructure from 2014 to 2024. CiteSpace was used for bibliometric analysis of publication volume, keyword co-occurrence, clustering, emergence and cited literature co-occurrence analysis.Results:A total of 5 389 English papers and 1 659 Chinese papers were included, showing an increasing publication trend. High-frequency Chinese keywords included prediction, influencing factor, and machine learning, while English keywords frequently used were machine learning, prediction, and artificial intelligence. The Chinese keywords formed 14 clusters such as epidemiological characteristic, dietary pattern, and elderly individual, and the English keywords formed 21 clusters including prediction model, risk factor, and adult. In international studies, health policy, COVID-19, and digital health were the emerging frontier keywords. Eleven core papers were selected, covering key areas like traffic accident risk assessment, public health big data application, and deep learning in medical diagnosis.Conclusions:This study systematically summarized the research hotspots and development trends of AI applications in epidemiology over the past decade by using bibliometric methods, which indicated that current AI-based epidemiological studies are still in the exploratory phase, with the coexisting of both advantages and challenges. Continued attention should be paid to the future development of this field.