Hepatitis C virus (HCV) infection remains a critical public health issue worldwide. Direct-acting antivirals (DAAs) have revolutionized the treatment of hepatitis C. However, real-world elimination efforts are hindered by barriers in diagnosis, treatment access, and follow-up. Embedding patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs) into routine care may improve service delivery. This study evaluates clinical and patient-reported outcomes in the HCV care cascade in Tuscany (Italy), offering insights into how health service organization affects effectiveness, equity, and patient experience. We conducted a multicenter, longitudinal, prospective study on 953 adults with HCV chronic infection. These adults were treated between 2021 and 2023 in seven prescribing centers in Tuscany. Clinical data included demographics, comorbidities, fibrosis staging, virological response (SVR12), and loss to follow-up (LTFU). PROMs and PREMs were collected at baseline (T0), 3 months post-treatment (T1), and 6 months after T1 (T2). We used the SF-12 tool and custom surveys. Patients were stratified by referral source (GPs, harm reduction services/prison, specialists). Clinical and questionnaire data were analyzed separately. Statistical analyses included ANOVA, Chi-square, Kruskal-Wallis, Cochran's Q, Friedman, and repeated measures ANOVA with Bonferroni corrections. Significance was set at p ≤ 0.05. DAA therapy achieved high clinical efficacy: SVR12 was reached in 93.4% of patients. The rate rose to 98.6% when excluding those lost to follow-up. Patients referred by harm reduction/prison services were younger and mostly male. They had higher psychiatric comorbidities and risk behaviors. PROMs demonstrated significant improvements in perceived physical and emotional health following treatment, particularly among individuals referred by GPs and specialists. PREMs revealed increasing satisfaction with referring doctors over time. Satisfaction with specialist care remained high and stable. Referral pathways markedly influenced patient profiles and reported outcomes. There were notable disparities in experience and quality-of-life indicators. This study highlights the importance of integrating clinical and patient-reported data in monitoring HCV care. High SVR rates confirm the effectiveness of DAAs, while PROMs and PREMs provide valuable insight into patient engagement and equity of access. Stratified analyses reveal the need for tailored approaches across care pathways, and high-risk populations require special attention. Embedding patient voice in evaluation fosters a more responsive, people-centered health system, advancing progress toward HCV elimination.

Hospital accreditation is intended to support hospitals in their quality improvement efforts. Accredited hospitals are expected to establish, maintain, and enhance their patient safety management systems. This study aimed to evaluate the characteristics of patient safety management systems in Japanese accredited hospitals and to investigate whether there were differences in patient safety management systems between accredited and non-accredited acute care hospitals. We conducted a secondary analysis of questionnaire survey data on patient safety management systems, collected every two years from 2015 to 2023 (five surveys in total). Approximately 3,000 hospitals were stratified by hospital size and randomly selected nationwide for each survey. From the survey items, we analysed seven methods related to identification of medical accidents and near-misses: Careful examination for all inpatient deaths, Complaints or questions from patients/family members, In-hospital reporting system of complications and comorbidities, Chart review, Occurrence review, Clinical conference, Morbidity and mortality conference (M&M conference), as well as the experience of serious medical accidents within the past three years. Trend comparisons between accredited and non-accredited hospitals were conducted using the Jonckheere-Terpstra test. Logistic regression analysis was also performed to assess the association between accreditation status and the experience of serious medical accidents. Across all seven methods for identifying medical accidents and near-misses, the implementation rates were consistently higher in accredited hospitals than in non-accredited ones. Trend analysis showed increasing trends in the implementation of Careful examination for all inpatient deaths, Complaints or questions from patients/family members, In-hospital reporting system of complications and comorbidities, Occurrence review, Clinical conference, and M&M conferences. The proportion of hospitals that experienced serious medical accidents also increased among accredited hospitals. Logistic regression analysis revealed that accredited hospitals were more likely to report serious medical accidents, which may be associated with differences in detection and reporting systems. The increasing trend in the use of several mechanisms suggests an association between accreditation and the presence of structured patient safety management systems. The enhanced identification ability may explain the observed increase in the reported experience of serious medical accidents. These findings suggest that accredited hospitals may be characterised by more developed or formalised approaches to identifying medical accidents and near-misses, as well as by differences in reporting culture. Further research is needed to clarify the mechanisms linking accreditation to improved safety outcomes.

Medical advances have markedly improved survival among patients with cardiovascular disease (CVD). Psychosocial and behavioral factors, including stress and physical activity, are increasingly recognized as key determinants of outcomes in this population. We investigated the association between existing CVD and perceived stress and examined how this association was modulated by physical activity levels in a Danish cohort, accounting for physiological and psychosocial conditions. We analyzed data on 102,650 Danish residents aged 50-65 years by linking self-reported questionnaire data from the Better Health in Late Life cohort with national health registry data. CVD was identified using hospital discharge diagnoses. Stress was measured with the ten-item Perceived Stress Scale. Physical activity was assessed with a pilot-tested questionnaire. Seven percent of participants had CVD. Compared with those without CVD, they had a higher prevalence of moderate-to-high perceived stress (42% vs. 33%) and a slightly lower prevalence of meeting guideline-recommended physical activity (42% vs. 45%). High stress was associated with physical activity not meeting the recommended level in both groups; however, this association was stronger among those with vs. without CVD (prevalence ratio 2.01 (95% confidence interval [CI]: 1.89-2.14) versus 1.52 (95% CI: 1.48-1.56)). We observed similar associations between stress and cardiovascular risk factors, comorbidity, psychosocial factors, and socioeconomic status. The association between stress and low physical activity persisted after stratification by these conditions. In conclusion, high stress was most strongly associated with low physical activity among adults with CVD. The underlying mechanisms seem to extend beyond physiological pathways and need further investigation.

BackgroundMammography screening uptake in Singapore remains below 40% despite campaigns and subsidies. Natural language processing (NLP) can extract nuanced attitudes from free text that fixed response options miss, revealing latent factors influencing breast cancer (BC) screening behavior.ObjectiveThis study characterized women’s attitudes toward mammography using mixed methods data, examined associations between BC awareness and screening willingness, and identified barriers and facilitators through NLP of free-text responses.MethodsWe conducted a cross-sectional study within the Breast Screening Tailored for Her multicenter cohort in Singapore (October 2021-December 2023). In total, 4169 women aged 35‐59 years (median 48, IQR 43‐54) were recruited via convenience sampling (3 hospitals and 2 polyclinics). Participants completed online structured questionnaires on demographics and screening history, then a BC education quiz with feedback. Participants answering >80% correctly were classified as “BC-aware.” Posteducation, participants reported screening willingness (motivated or neutral) with optional free-text explanations. Logistic regression models (adjusted for study site, age, ethnicity, marital status, housing, and education) examined the associations with willingness. For 3819 English-language respondents, biterm topic modeling identified themes and sentiment analysis quantified emotional tone. Statistical significance: α=.05.ResultsOverall, 79% (3287/4169) were BC-aware, and 94% (3908/4169) reported increased motivation posteducation. BC-aware women had higher screening motivation than BC-unaware women (adjusted odds ratio [aOR] 2.88, 95% CI 2.19‐3.80; P<.001). Motivation was higher among those with larger public housing (OR 1.81, 95% CI 1.30‐2.50; P<.001) and private housing vs 1‐3 room units (OR 2.69, 95% CI 1.75‐4.13; P<.001), married vs not separated, divorced, or widowed (OR 2.38 [inverse of 0.42], 95% CI 1.75‐3.13; P<.001), and prior screening attendance (OR 3.49, 95% CI 2.71‐4.50; P<.001). Women who disagreed that mammography was expensive had higher motivation (aOR 1.94, 95% CI 1.50‐2.50; P<.001). Among 3819 English respondents, 94% (3579/3819) were motivated and 6% (240/3819) neutral. Free-text responses came from 34% (1220/3579) of motivated and 64% (153/240) of neutral participants. Biterm topic modeling revealed motivated participants emphasized early detection benefits, health awareness, BC risk, and logistics; neutral participants focused on mammography pain experiences and cost barriers. Mean sentiment was 0.207 (range: −1.00 to 1.65), with motivated participants displaying more positive sentiments than neutral participants (linear regression, P<.001). Identical words carried different emotional tones across subgroups: “health” had positive sentiment among motivated participants (mean difference, Welch t tests P<.05) but negative sentiment among neutral participants. Word frequency analysis showed motivated participants used positive-sentiment words (“better,” “cure,” and “prevention”). Neutral participants emphasized negative words (“painful” and “uncomfortable”).ConclusionsIntegrating quantitative surveys with NLP revealed that the same screening concepts are emotionally framed differently by motivated vs neutral women, a finding missed by knowledge- or intent-focused approaches alone. In practice, these findings support the need for emotionally tailored BC education and prevention strategies.

Our study introduced the 2023 UK Biobank sleep questionnaire and described variation in sleep health dimensions and prevalence of disordered sleep. A questionnaire comprising validated measures and bespoke items was developed to capture key self-reported domains of sleep health and symptoms of sleep disorders. We quantified cohort prevalence of operationally defined sleep disorders and assessed patterning of sleep health dimensions across key sociodemographic and clinically relevant variables. 183,704 individuals completed at least one module of the questionnaire after email invitation (representing 56% of those with an active email address), and an additional 1,352 individuals completed via the participant website. A total of 185,056 individuals were included in the analysis. Respondents were predominately from a White ethnic background (96.8%), had a mean age of 69.9 (SD, 7.5) years, 57.9% were female, and 25.5% were in employment. Compared to non-respondents, respondents were more likely to be female, tended to be better educated, healthier, and exhibit lower levels of socioeconomic deprivation, although baseline sleep variables were similar between respondents and non-respondents. Around 40% of respondents reported sleep duration less than 7 hours and 49% reported poor sleep quality (Pittsburgh Sleep Quality Index > 5). Approximately one-quarter (25.2%) met criteria for at least one operationally defined sleep disorder, with insomnia being the most common (14.4%) followed by obstructive sleep apnoea (8.0%), restless legs syndrome (4.1%), and frequent nightmares (3.7%). Sleep disorders were associated with higher levels of anxiety, depression, fatigue, and cognitive complaints. Poor sleep quality and operationally defined sleep disorders are common in the UK Biobank cohort. Sleep questionnaire data can now be integrated with a range of biomedical information to advance understanding of sleep. Statement of significance A comprehensive sleep questionnaire was introduced to the UK Biobank, with over 185,000 participants providing data. Overall, respondents reported relatively poor sleep quality; 40% reported sleep duration less than 7 hours, and 25% met criteria for at least one sleep disorder. Enhanced assessment of sleep in UK Biobank now enables integration with extensive biomedical data, including genetic, wearable, imaging, lifestyle, biomarker, and electronic health record data, offering opportunities to investigate the biological and environmental factors that influence sleep and circadian systems, and their impact on health.

The tourism industry is undergoing rapid digital transformation, with augmented reality (AR) technologies increasingly integrated into tourism services. AR real-scene maps enhance the tourism experience, yet existing research focuses more on technology and less on user adoption. Based on TAM, this study introduces system quality, information quality, immersion, and perceived enjoyment to examine tourists’ usage intention. Using questionnaire data from 322 respondents and SEM, findings show significant effects among key variables. Gender, age, and usage experience also play moderating roles. This study offers theoretical and practical implications for AR map design and tourism experience optimization.

Objective. To compare the efficacy of various clinical monitoring programs, including a remote monitoring system using automated interactive questionnaire. Materials and methods. The study included 254 patients with COPD, divided into 3 dynamic monitoring groups: the main group of patients (n=77) used a remote monitoring system activities, including an interactive questionnaire; the comparison group (n=75) who used a remote monitoring system activities limited to the assessment of a number of objective parameters: a modified 6-minute walk test (6-MWT test, in steps), daily physical activity (number of steps per day) and peak expiratory flow rate, but the interactive questionnaire was not applied in this group; the control group (n=102), received standard follow-up according to routine clinical practice without remote monitoring in compliance with the Order No. 168n of the Russian Ministry of Health (dated March 15, 2022) “On approval of the Procedure for Dispensary Observation of Adults”. Results. In the main group, where remote monitoring of interactive questionnaire data was combined with the assessment of objective disease criteria, a significant proportion of patient transition from group “E” to group “B” (a course with rare exacerbations) was revealed. In the comparison group, no significant disease rephenotyping was observed. Conversely, in the control group, a significant proportion of patients transitioned from groups “A” and “B” to group “E” (a course with frequent exacerbations). Over the 12 month follow-up period, treatment in the main group was aligned with clinical guidelines; in the comparison group, a less significant redistribution of treatment programs was observed; in the control group, the rate of treatment regimen compliance with clinical guidelines after one year did not exceed 40 %. Conclusions. An approach combining the assessment of subjective (using an interactive questionnaire) and objective criteria for remote monitoring, improves the quality of clinical follow-up in COPD patients, allowing disease rephenotyping, timely detection of disease exacerbations, and achieving symptom control through the prompt adjustment of maintenance therapy, when required.

Background: Emergency department nurses face a high risk of burnout due to high-intensity workloads, long working hours and workplace violence, which is associated with the quality of care and increases nurse turnover. Traditional interventions often lack dynamic adjustment capabilities. Methods: This narrative review integrates research on burnout monitoring and prediction, system-level interventions, and individual-level interventions. Results: Machine learning models trained on questionnaire data can tell whether it is a risk when burning out. Wearable devices enable real-time monitoring to prevent burnout at the early stage of detection. But the problem also needs to be resolved with attention to compliance and measurement sensitivity. The problem also needs to be addressed in terms of compliance and recognition of sensitivity. AI-assist scheduling and prediction make system less unequal and lessen work burden. In terms of individuals, modules and digital psychological interventions through the just in time adaptive intervention framework with algorithm matching and the hybrid model of self-selection is to improve the user's interaction. Conclusions: AI-driven interventions for burnout among emergency department nurses provide evidence for the feasibility of proactive, closed-loop intervention programs. It is better to pay more attention to predicting the results of prediction. Translating this into concrete actions and sustaining it over the long term can effectively reduce burnout and maintain a good nursing turnover rate.

AimTo describe headache phenotypes and treatment patterns in adults with moderate to severe post-traumatic headache and a high burden of other post-concussion symptoms two to four months after mild traumatic brain injury.MethodsA cross-sectional study was conducted using patient-reported questionnaire data from 218 individuals, aged 18-60 years, with moderate to severe post-traumatic headache, a headache duration between two and four months, and a high burden of other post-concussion symptoms. The study was based on data from a cohort study with a nested randomized controlled trial called Get Going After Concussion 2.0. Individuals were recruited from emergency departments and general practitioners in Central Denmark Region (1.4 million inhabitants). Phenotyping was performed using an algorithm linking the questionnaire data to diagnostic criteria of primary headache disorders.ResultsAmong the 218 individuals, 62% had acute- and 38% had persistent post-traumatic headache. A migraine-like or probable migraine-like phenotype was identified in 88% of the individuals with no significant differences in phenotype distribution between the acute- and persistent headache subgroups (p = 0.414) or between sexes (p = 0.994). Headache frequency of ≥15 days per month was reported by 78% and average headache intensity was moderate or severe in 94% of all individuals. A pressing and/or tightening headache quality was the most frequent (85%). No significant differences were found between the acute- and persistent headache subgroups regarding pharmacological treatment use or efficacy (p = 0.776). Non-opioid analgesics had been used by 88% of all individuals and 81% were still using them; 52% reported lack of efficacy. Triptans had been or were currently used by 4% of all individuals. Preventive medication, used by 7% of the study population, was reported to have lack of efficacy but all continued treatment. Probable medication overuse headache was found in 27% of the individuals. Overall, 33% had tried two or more non-pharmacological treatments, and 6% had tried between four and six.ConclusionA migraine-like phenotype was the most common presentation in adults with moderate to severe post-traumatic headache and a high symptom burden of other post-concussion symptoms two to four months after trauma. Overall, limited effect of both pharmacological and non-pharmacological treatment was reported. Preventive medication and triptans were used by very few individuals, despite being recommended in current treatment guidelines. The study population exhibited a high health care-seeking behavior, and a large proportion had probable medication overuse headache. Consequently, further insights into disease-specific treatment approaches of post-traumatic headache is needed.

Towards a representative and affordable national survey on radon in dwellings: a new methodological approach with findings from Italy.

Disrupting traditional silos and boundary crossing: Interprofessional precepting in a fourth-year pharmacy rotation.

Abstract Previous studies have emphasized that individuals play important roles in facilitating organizational unlearning; however, little is known about how leadership promotes organizational unlearning. From the perspective of routine dynamics, this study explores the effects of managers' behaviors on unlearning at the department level. Open‐ended questionnaire survey data were derived from 385 managers of intellectual property (IP) departments in Japan. A mixed‐methods approach combining the grounded theory approach (GTA) with content analysis was used. The GTA generated three types of departmental unlearning (improving efficiency, networking, and strategizing) and five types of managers' behaviors (initiation, planning, upward communication, horizontal communication, and implementation). Improving efficiency is regarded as operative or technical unlearning, whereas networking and strategizing are regarded as reinventive or adaptive unlearning. Based on content analysis, binary logistic regression analyses showed that managers' upward and horizontal communication were positively associated with networking and strategizing in IP departments, whereas implementation was positively associated with strategizing. This study provides new insights into the literature by identifying the specific managers' behaviors in promoting strategic unlearning in departments.

Integrating artificial intelligence to enhance inclusive education for students with special needs in Jordan and the UAE: Perspectives of teachers and parents.

The association between flavonoid intake and first clinical diagnosis of central nervous system demyelination: a case-control study.

Mobile health (mHealth) technology has been utilized to offer self-management tools to people with pain, including symptom tracking. Existing mobile tracking applications (apps) for chronic pain management have demonstrated reliability, feasibility, improved coping, and reduced health care utilization. Unfortunately, adherence in using a pain app can be problematic with many not using or discontinuing its use early. The current study aimed to investigate the impact that pain self-efficacy, pain conceptualizations, and patient perception of pain care providers, have on engagement with a mobile pain tracking app. Seventy-six (N=76) individuals with chronic pain downloaded a pain app and completed questionnaires assessing their pain and use of a pain app 3 months after they had downloaded the app. Associations with engagement with the app, defined as the number of daily diaries completed, and demographic and self-report questionnaire data were examined. Results showed that engagement with the app was unrelated to self-efficacy and pain conceptualization but significantly related to positive perceptions of their pain care providers. Patients with more severe pain were found to have lower self-efficacy, less understanding of the biopsychosocial model of pain, and lower satisfaction with their involvement in their pain care decisions. Surprisingly, those who engaged more with the app demonstrated lower self-efficacy as compared with those who used the pain app less. These findings highlight the importance of the patient-provider relationship in engaging with mHealth technology for pain management. Results further imply that longer-term use of mHealth tools may not be perceived as adaptive or clinically helpful for certain individuals.

Non-communicable diseases (NCDs), constitute a major global public health challenge and represent the primary causes of mortality and morbidity worldwide. This study aims to estimate Quality-Adjusted Life Year (QALY) losses associated with six NCDs among Iranian adults. This study quantified QALY losses associated with six NCDs, including asthma, ischemic heart disease (IHD), stroke, hypertension, diabetes mellitus, and high cholesterol. Health-related quality of life (HRQoL) scores were derived from EQ-5D-3L questionnaire data collected in 2021 from a nationally representative sample of 27,576 participants. Morbidity prevalence was obtained from the same survey, while mortality data were sourced from the Global Burden of Disease (GBD) study. Total QALY loss for each condition was calculated by summing losses attributable to both morbidity and mortality. Women experienced a greater decline in HRQoL than men across all conditions. The highest disutilities were observed for stroke, IHD, and asthma in women, and for stroke, asthma, and IHD in men. The highest QALY losses were associated with hypertension (1,399,097), IHD (1,123,053), and high cholesterol (749,136). Diabetes mellitus accounted for 428,163 QALYs lost followed by Stroke (373,365) and asthma (215,498). Given the substantial health burden posed by NCDs, there is an urgent need for prevention and management strategies that are both evidence-based and gender-sensitive. Strengthening national policies aimed at reducing the NCDs burden will not only enhance population health outcomes but also generate significant economic returns.

The rapid digital transformation in the Industry 4.0 era has reshaped vocational education, emphasizing the need for graduates who possess strong English communication skills aligned with global business demands. This study investigates students’ perceptions of using Free4Talk, an online language exchange platform, to enhance English for Business Communication among vocational students. Employing a mixed method design, the research collected quantitative data through a five point Likert scale questionnaire and qualitative data through semi-structured interviews with 20 Business Administration students at Kupang State Polytechnic. Descriptive statistics were used to analyze the questionnaire, while thematic analysis was applied to interpret interview responses. The findings revealed that students hold positive perceptions of Free4Talk across four key dimensions: usability and accessibility (M=3.73), motivation and engagement (M=3.59), language and business communication skills (M=3.45), and intercultural communication competence (M=3.43) all categorized as “high.” The results indicate that Free4Talk effectively fosters speaking fluency, confidence, and intercultural awareness through authentic, real time interaction with global users. Students valued its accessibility, user friendly design, and motivational features that encouraged frequent speaking practice beyond the classroom. In conclusion, integrating Free4Talk into vocational English instruction bridges academic and workplace communication needs, supporting the development of globally competent communicators. It is recommended that future research employ experimental or longitudinal approaches to examine measurable gains in fluency, accuracy, and pragmatic competence, as well as to explore the integration of teacher guidance and feedback mechanisms in technology-mediated language learning

Teachers&amp;rsquo; emotions play a critical role in shaping the classroom atmosphere and influencing students&amp;rsquo; learning experiences. In many college English classrooms across China, students often exhibit characteristically low participation and weak motivation. This study examines the dynamic mechanisms through which teachers&amp;rsquo; emotions affect students&amp;rsquo; learning engagement&amp;mdash;encompassing behavioral, emotional, and cognitive dimensions&amp;mdash;as well as their intrinsic and extrinsic motivation. Employing a mixed-methods approach, the research analyzed questionnaire data from 107 students across five Chinese universities, supplemented by semi-structured interviews with three instructors and twelve students. Findings reveal that positive teacher emotions, such as enthusiasm and encouragement, significantly enhance classroom interaction and promote sustained independent learning, while also serving as a positive predictor of intrinsic motivation. Conversely, negative emotions, including anxiety and apathy, were associated with classroom avoidance behaviors among 93.02% of respondents and were linked to reduced stability in goal orientation. Mediation analysis further supports the proposed pathway of &amp;ldquo;teachers&amp;rsquo; emotions &amp;rarr; learning engagement &amp;rarr; learning motivation,&amp;rdquo; highlighting the mediating role of teacher-student emotional connection. These findings contribute to understanding affective transmission in second language acquisition and offer an empirical basis for developing teacher emotional literacy training and designing emotionally supportive pedagogical strategies, with practical implications for enhancing the ecology of foreign language instruction.

ABSTRACT Given the pivotal role of learner engagement in shaping second language (L2) development, identifying its motivational underpinnings is of growing interest. However, research has largely focused on isolated variables, offering limited insight into how multiple learner factors interact. This study addresses this gap by examining how three motivational constructs – basic psychological needs (autonomy, competence, and relatedness), language mindset (growth vs. fixed), and academic buoyancy – independently and configurationally influence L2 learner engagement. Drawing on both multiple regression and fuzzy-set Qualitative Comparative Analysis (fsQCA), this study analyzed questionnaire data from 539 college students who learn English as a Foreign Language. The results indicated that 1) competence, relatedness, growth mindset, and buoyancy positively predicted engagement, while autonomy and fixed mindset did not show statistical influence; 2) no single factor emerged as a necessary or sufficient condition for sustained engagement; 3) five distinct motivational configurations were found to lead to high engagement, showing the interactive nature of these factors and complexity of engagement. This study expands our existing knowledge about the underlying mechanism of how diverse psychological resources shape engagement by the configurational methods. These results offered practical and tailored pedagogical implications for enhancing students’ engagement during L2 learning process.

The problem of mental health concerns among university students is growing more and more noticeable, and early identification of mental health signs is crucial for intervention. Current assessment methods rely on questionnaire data, which suffer from inefficiencies in data statistics. Taking Rednote as a case study, this research employs web crawling technology to gather diverse user information. By utilizing data mining techniques, linguistic, emotional, and behavioral features are extracted, and a depression knowledge graph is constructed to model the deep-level associations among these features. Additionally, the study selects a Convolutional Neural Network (CNN) and a Gated Recurrent Unit (GRU) to develop a predictive model that integrates textual features with user characteristics, ultimately yielding mental health predictions. The results indicate that the predictive model achieves an accuracy of 87.3% and an F1 score of 86.5%. Without the knowledge graph, the accuracy drops to 81.2%, representing a 6.1% decrease compared to the predictive model, demonstrating that the knowledge graph can effectively identify key depression-related pathways. The F1 score for the combined CNN and GRU model reaches 98.5%, showing a 6.3% improvement over the GRU alone. This research provides a feasible approach for social media data-driven mental health monitoring, with the application of the knowledge graph enhancing model interpretability and aiding in the development of precise psychological intervention strategies. It offers an automated screening tool for college psychological counseling services and provides data support for depression prevention research in the realm of public health.