To explore the potential categories and influencing factors of limb swelling after total knee arthroplasty(TKA).

A total of 239 patients undergoing TKA in the department of orthopedic surgery of a tertiary grade A hospital in Jiangsu province from June 2023 to May 2025 were selected as survey participants. Baseline surveys were conducted within one week postoperatively using the General Information Questionnaire,Limb Swelling Circumference Assessment Form, Pictorial Knee Self-Assessment Questionnaire,Orthopedic Patient Functional Exercise Adherence Scale,and Social Support Rating Scale.Limb swelling severity was assessed using the Limb Swelling Circumference Assessment Form on postoperative days 1,3,and 7,as well as at 1,3,and 6 months after discharge.Growth mixture modeling was used to identify potential categories of limb swelling after TKA,and logistic regression analysis was performed to examine influencing factors.

A total of 206 patients(86.19%) completed the follow-up. Three potential categories of limb swelling were identified:high-level swelling with a high plateau(30.58%),moderate-level swelling with fluctuation (35.44%),and low-level swelling with a decreasing trend(33.98%).Logistic regression results showed that inflammation level,pain severity,knee joint function,comorbidities,functional exercise adherence,and social support were significant influencing factors of the potential categories of limb swelling after TKA(P<0.05).

Limb swelling after TKA exhibits population heterogeneity.Based on these findings,healthcare providers can implement stratified assessment and personalized interventions for postoperative swelling to improve patient outcomes in swelling rehabilitation.

To explore the risk factors for hematochezia in infants with cow's milk protein allergy(CMPA) and to construct and to validate a prediction model.

A retrospective study was conducted.A total of 408 infants with CMPA who visited the pediatric or child health care outpatient departments of tertiary hospitals from January to December 2024 were included as the modeling group.Another 175 infants with CMPA were included as the validation group.Participants were divided into a hematochezia group and a non-hematochezia group based on whether hematochezia occurred.Characteristic variables were selected using a combination of LASSO regression and the Boruta algorithm. Five machine learning algorithms-random forest,support vector machine,extreme gradient boosting(XGBoost),decision tree,and Logistic regression,were applied to construct prediction models.The area under the receiver operating characteristic curve(AUC) from internal validation was used as the primary criterion for selecting the optimal model.The optimal model was then subjected to external validation and interpretability analysis.

Feeding intolerance,age in months,eczema,early-life antibiotic exposure,formula feeding,maternal consumption of allergenic foods during lactation,maternal history of allergy,and first feeding(formula) were identified as independent risk factors for hematochezia in infants with CMPA.The XGBoost model demonstrated the best predictive performance,with AUC values of 0.841 in the modeling group and 0.821 in the validation group.SHAP values of the XGBoost model revealed that feeding intolerance,age in months,and eczema were the top three important predictive factors.

The XGBoost model shows good performance in predicting the risk of hematochezia in infants with CMPA. It could serve as a reference for healthcare professionals to identify high-risk infants early.

To construct a dynamic adjustment theoretical framework of childbirth fear among nulliparous women by using grounded theory.

A total of 35 nulliparous women were recruited by using purposive and theoretical sampling and participated in semi-structured in-depth interviews,with data analyzed by using grounded theory.

A total of 10 first-order categories were identified,including psychological factors,physiological factors,social factors,antecedent stimuli,risk perception,need fulfillment,cognitive restructuring,belief enhancement,coping behaviors,and control.These were further integrated into four second-order categories:causes of childbirth fear,adjustment motivation,adjustment strategies,and adjustment outcomes.The core category,termed the dynamic adjustment process of childbirth fear,was identified,based on which a theoretical framework describing nulliparous women's dynamic adjustment to childbirth fear was developed.

The theoretical framework of dynamic adjustment of childbirth fear among nulliparous women could assist obstetric healthcare professionals in the early identification of childbirth fear and provide theoretical basis for developing targeted nursing interventions and optimizing maternity care processes.

Taking the hospital needlestick injury report as the application scenario,to construct a comprehensive assessment framework of "theme quality-interpretability-action transformation-technical characteristics" to compare the potential Dirichlet scores,the differences between the two topic modeling methods,LDA and the Deep Learning Topic Framework Based on Pre-trained Language Models(BERTopic),in risk identification and management decision support,and input the theme-driven intervention suggestions.

A total of 160 needlestick injury reports were collected from a tertiary hospital between January 2018 and September 2025.After preprocessing including Chinese word segmentation,stop-word removal,and synonym merging,LDA and BERTopic models were trained separately.Model comparison was conducted using topic coherence,expert blind evaluation,and technical indicators such as training time,parameter sensitivity,and degree of automation.The resulting topics were then mapped onto a tiered intervention framework to formulate actionable countermeasures.

Both models identified key risk scenarios including blood specimen collection, needle withdrawal/needle handling,and complex procedures in special departments.LDA generated five macro-level topics,suitable for rapid overview and routine monitoring.BERTopic generated four semantically coherent topics and one noise cluster, which more clearly delineated actionable scenarios such as blood collection chain risks among intern nurses,systemic failure in post-needle processing,and exposure in high-infection-risk cases.The topic coherence scores for BERTopic and LDA were 0.587 and 0.452, respectively.The overall expert rating scores were 4.555 and 3.944,respectively.Training times were 182 seconds and 45 seconds,respectively.

BERTopic demonstrates superior advantages in semantic depth and action translatability, making it suitable for root cause analysis and targeted improvement initiatives.LDA is better suited for rapid daily screening and trend monitoring.This study recommends a tiered application strategy of "routine LDA monitoring plus periodic BERTopic deep dives" in hospitals.The topic results can be transformed into structured interventions and evaluation indicators targeting populations,workflows,and devices,thereby forming a closed-loop management cycle of risk identification-intervention-review.

To translate the Pregnancy Eating Attitudes Questionnaire(PEA-Q) into Chinese and test its reliability and validity.

Following Brislin's translation model,the English version of PEA-Q was forward-translated,back-translated,cross-culturally adapted,and pilot-tested to develop the Chinese version of PEA-Q.From June to October 2025,1 032 pregnant women were recruited from one tertiary hospital and three community health centers in Changzhou.Classical test theory and item analysis were used to evaluate reliability,validity,measurement invariance,and item performance.

The Chinese version of PEA-Q consisted of 11 items across three dimensions:permissive eating attitude,self-regulated eating attitude,and mildly restrained eating attitude.The scale-level content validity index was 1.000,and the item-level content validity index was 1.000.Exploratory factor analysis extracted three common factors,with a cumulative variance contribution rate of 78.075%.Confirmatory factor analysis indicated a good fit for the three-factor model.The overall Cronbach's α coefficient was 0.895.The questionnaire demonstrated measurement invariance across different gestational weeks and age groups.Item discrimination ranged from 1.59 to 2.80,and item difficulty ranged from -2.02 to 2.17.

The Chinese version of PEA-Q shows good reliability,validity,and cross-group measurement invariance.Its items exhibit strong measurement performance. It is a reliable tool for assessing eating attitudes during pregnancy.

To compare the psychometric properties,feasibility,and clinical utility of the Pain Intensity Measure for Persons with Dementia(PIMD) and the Mobilization-Observation-Behaviour-Intensity-Dementia Pain Scale(MOBID-2) for pain assessment in hospitalized older adults with cognitive impairment.

A total of 134 hospitalized older adults with cognitive impairment from a tertiary grade A hospital in Guangdong province were enrolled.Using the Chinese version of the Pain Assessment in Advanced Dementia Scale(PAINAD) as the criterion,patient's self-reported Numeric Rating Scale pain scores were obtained,followed by pain assessments at rest and during activity using the three scales(PAINAD,PIMD,and MOBID-2) in random order.The internal consistency,inter-rater reliability,criterion validity,construct validity,and pain screening ability were compared between PIMD and MOBID-2.Additionally,22 clinical nurses from six hospitals in Guangdong Province evaluated the feasibility and clinical utility of the two scales.

At rest and during activity,the Cronbach's alpha coefficients of PIMD were 0.821 and 0.889,and the inter-rater reliability coefficients were 0.847 and 0.839,respectively.Exploratory factor analysis extracted two common factors for the PIMD in both states,with cumulative variance contribution rates of 71.047% and 77.402%,respectively.Regarding criterion validity,the correlation coefficients between the PIMD and the PAINAD were 0.996 at rest and 0.954 during activity.During activity,the Cronbach's alpha coefficient of the MOBID-2 was 0.781,and the inter-rater reliability coefficient was 0.775;exploratory factor analysis extracted four common factors with a cumulative variance contribution rate of 67.229%;the correlation coefficient between the MOBID-2 and the PAINAD was 0.883.The area under the receiver operating characteristic curve(AUC) for the PIMD and the MOBID-2 were 0.984 and 0.930,with sensitivities of 91.9% and 71.2%,and specificities of 95.7% and 100.0%,respectively.The Kappa values for agreement among the three scales were all>0.4,and the pain detection rates ranged from 59.0% to 82.8%.The mean feasibility and utility scores for the PIMD and the MOBID-2 were 4.2 and 4.2,3.7 and 4.2,respectively.

The PIMD demonstrates good reliability and validity,while MOBID-2 shows good reliability but moderate validity.Both scales can be used for pain assessment in hospitalized older adults with cognitive impairment.However,considering the psychometric properties,feasibility,and clinical utility of each scale,PIMD is recommended as the priority choice.

To develop a journey map of home health management for patients with tracheoesophageal fistula after "three-tube" interventional procedure,systematically analyze their core experiences and needs at different stages,and propose targeted nursing strategies accordingly.

From March to August 2025,patients who underwent "three-tube" procedure for tracheoesophageal fistula and their caregivers were selected from the interventional radiology department of a tertiary general hospital in Zhengzhou city for readmission follow-up as study subjects.Semi-structured in-depth interviews were conducted,and the interview data were analyzed using Colaizzi's seven-step analysis method to develop a patient journey map.

A total of 26 sub-themes were summarized from tasks,emotions,pain points dimensions,forming a journey map covering the early discharge period,transitional adaptation period,long-term tube-living period,and rehabilitation sprint period.

The home health management needs of patients with tracheoesophageal fistula after "three-tube" interventional procedure present stage-specific and multidimensional characteristics,with prominent needs for caregiving skill support,psychosocial adaptation,and swallowing safety reconstruction.In the future,phased and systematic transitional nursing strategies can be developed and implemented based on the constructed health management journey map.

To explore multifactor combination patterns of unplanned extubation using association rule mining and to construct a Bayesian network predictive model.

A retrospective case-control study was conducted.Clinical data were collected from 133 patients with unplanned extubation and 350 control patients at a tertiary grade A hospital in Guizhou province between January 2013 and December 2023.The Apriori algorithm was used to mine association rules and screen strong association rules. Meanwhile,the max-min hill-climbing(MMHC) algorithm was employed to construct a Bayesian network predictive model,which was used to analyze causal relationships and risk probabilities among various factors.

A total of 20 strong association rules were identified. Factors such as night shift,years of nursing experience,caregiver responsibility lacking,omission of catheter assessment,pain,agitation,restraint status,catheter retention time,and age were closely associated with unplanned extubation.The Bayesian network model consisted of 10 nodes and 14 directed edges,and the incidence of unplanned extubation under high-risk combinations could reach 90%.

Unplanned extubation results from the interaction of multiple factors,with omission of assessment during handover representing a core risk link.The predictive model constructed in this study demonstrates strong risk identification and predictive capabilities,providing a quantitative tool for early warning and intervention in clinical settings.

To identify latent classes of fall risk perception in patients with posterior circulation ischemia(PCI),and to analyze their influencing factors,and to explore the impact of different classes on exercise self-efficacy.

A total of 200 PCI patients admitted to our hospital from August 2023 to July 2025 were enrolled.Surveys included a general information questionnaire,the Fall Risk Perception Scale,Pittsburgh Sleep Quality Index(PSQI),Berg Balance Scale(BBS),Exercise Self-Efficacy Scale(MSES),and Activities of Daily Living(ADL) scale.Latent profile analysis(LPA) was used to identify latent classes of fall risk perception and their influencing factors.The impact of these classes on exercise self-efficacy was further examined.

Fall risk perception in PCI patients fell into three latent classes:insufficient perception(78 cases,39.0%),moderate perception(92 cases,46.0%),and excessive perception(30 cases,15.0%).Multivariate logistic regression showed that PSQI score was a factor for the insufficient perception class(OR=1.553,P<0.001).BBS score,ADL score,MSES score,and fall history were factors for the excessive perception class(OR=0.840,0.882,0.933,and 2.945 respectively,P<0.05).Multiple linear regression showed that after controlling for confounders,the excessive perception class was a predictor of exercise self-efficacy score(P<0.05).

Fall risk perception in PCI patients showed population heterogeneity and it was influenced by sleep quality,balance function,and daily activity ability.The excessive perception class negatively predicted patients' exercise self-efficacy. Clinicians may consider identifying different patient classes and develope targeted intervention strategies.