To analyze the epidemiological characteristics and temporal trends in Lower Respiratory Infections（LRIs）mortality among population aged ≥60 years in China from 2010 to 2021, providing evidence for formulating and optimizing prevention and control strategies to reduce LRIs-related deaths.

LRIs mortality data for population ≥60 years were collected from the National Disease Surveillance System Cause of Death Surveillance Dataset（2010-2012）and the Chinese Cause of Death Surveillance Dataset（2013-2021）by sex, urban-rural residence, region, and age group. Mortality rates were standardized using the 2010 National Population Census data. Crude and age-standardized mortality rates were calculated to describe the mortality status of LRIs. Joinpoint 5.4.0.0 software was used to calculate the annual percent change（APC）and average annual percent change（AAPC）to analyze the temporal trends, and the chi-squared test was applied to compare mortality rates across groups.

From 2010 to 2021, a total of 255 385 LRIs deaths occurred among the monitored population aged ≥60 years in China, accounting for 87.65% of all LRIs deaths. The overall crude mortality rate was 57.45 per 100 000 population, with males（63.07/100 000）having a significantly higher mortality rate than females（52.15/100 000）, urban areas（77.95/100 000）higher than rural areas（47.04/100 000）, and the western region（66.10/100 000）higher than the eastern（59.83/100 000）and central（48.17/100 000）regions（all P<0.001）. The age-standardized mortality rate for LRIs among population aged ≥60 years showed an overall downward trend from 2010 to 2020（AAPC=-8.25%, P<0.001）. Similar trends were observed in males（AAPC=-7.77%, P<0.001）, females（AAPC=-9.04%, P<0.001）, urban areas（AAPC=-8.02%, P<0.001）, the eastern region（AAPC =-7.79%, P<0.001）, and the central region（AAPC=-8.58%, P<0.001）, while rural（AAPC=-8.04%, P<0.001）and western regions（AAPC=-8.09%, P<0.001）exhibited steady declines. Among the population aged ≥60 years, LRIs mortality increased significantly with age, while the magnitude of decline showed an upward trend. The AAPCs for the age groups 65-69, 70-74, 75-79, 80-84, and ≥85 years were -3.16%, -5.76%, -6.60%, -8.33%, and -9.90%, respectively（all P<0.05）.

From 2010 to 2021, the mortality rate of LRIs among population aged ≥60 years in China showed a downward trend, which increased significantly with age and showed marked disparities by sex, age, urban/rural residence, and geographic region.The elderly males in urban and western regions are high-risk groups for LRIs-related deaths and require focused attention and targeted control measures.

To systematically review the current research status, feasibility, and future prospects of using large language models（LLMs）for mental health interventions.

We searched literature published between January 2023 and March 2025 on LLM-based mental health interventions across databases including CNKI, Wanfang Data, VIP, PubMed, Scopus, and Web of Science. After screening, the Mixed Methods Appraisal Tool（MMAT）was used for quality assessment. Key data were extracted, including study characteristics, model types, and supported languages.

Twenty-one English-language studies were included. Twelve studies employed population-based intervention trials, investigating the effects of LLMs in reducing anxiety, identifying and intervening in cognitive distortions, measuring and fostering psychological resilience in children, alleviating depression and loneliness, and enhancing willingness to engage in mindfulness training. Results indicated that LLMs demonstrated some efficacy in these tasks. Another seven studies supported the feasibility of LLM-based mental health interventions in three key areas: psychological knowledge consultation, emotional support, and therapeutic assistance. However, challenges such as bias, cultural adaptability, and privacy concerns remain. Seven studies used English as the model language, while thirteen applied prompt engineering or instruction fine-tuning to adapt the models, with adjusted models showing improved performance in specific mental health tasks. Quality assessment revealed that existing studies were limited.

LLMs show promising potential in mental health promotion. Future research should refine model strategies to enhance task performance and incorporate rigorously designed empirical research to ensure scientifically robust efficacy evaluations.

To systematically evaluate the prevalence of SO in diabetic patients in China.

Search for literature on the prevalence of SO in Chinese diabetic patients in 8 databases including PubMed, Web of Science, Embase, etc, with the search timeframe spanning from the databases’ inception to February 2025. Statistical analysis was conducted using Stata 15.0.

This meta-analysis included 13 studies involving 5 554 diabetes patients. The pooled results demonstrated a 14.9%（95% CI: 11.0%–18.8%）prevalence of SO among diabetic patients in China. Subgroup analysis showed that the prevalence was higher in studies where the participants were aged 60 or above, male, from the southeastern region, used bioelectrical impedance analysis（BIA）, diagnosed sarcopenia based on only a single assessment of muscle mass or strength, and used waist circumference（WC）as the diagnostic criterion for obesity. Meta-regression showed that none of the results were statistically significant（P>0.05）, and no sources of heterogeneity were identified.

The prevalence of SO among diabetic patients in our country is relatively high. Therefore, it is necessary to actively carry out health education for diabetic patients, strengthen early screening and intervention management, in order to prevent and slow down the progression of the disease.

To identify causal genes associated with frailty, a significant global health issue whose genetic mechanisms remain poorly understood, by integrating multi-tissue gene expression data.

This study integrated expression quantitative trait loci（eQTL）data from 49 human tissues and employed summary data-based Mendelian randomization（SMR）to analyze causal associations between gene expression and frailty. Genetic colocalization of signals was used to validate significant genes, followed by sensitivity analysis using two-sample Mendelian randomization, with MR-Egger regression and Cochran’s Q test to control for pleiotropy and heterogeneity. Gene Ontology（GO）enrichment analysis was performed for genes identified through SMR, colocalization, and sensitivity analyses.

Twenty causal genes（80 significant associations）linked to frailty were identified across 27 tissues. The top five genes showing multi-tissue causal associations included LRPPRC, RBM6, TTC12, RNF123, and CPS1. The final 20 causal genes were primarily enriched in mitochondrial function regulation, nuclear-mitochondrial signaling, and redox metabolic processes.

The causal genes identified in this study may influence the occurrence and development of frailty through mitochondrial function and nuclear signaling pathways.

To investigate the correlation between the C-reactive protein-albumin-lymphocyte（CALLY）index and chronic obstructive pulmonary disease（COPD）.

Using data from two cycles of the National Health and Nutrition Examination Survey（NHANES）in the United States from 2015 to March 2020, we conducted multivariate logistic regression, subgroup analysis, and restricted cubic spline analysis to explore the relationship between Log₂CALLY and COPD. The predictive accuracy of Log₂CALLY for COPD risk was assessed using receiver operating characteristic（ROC）curve analysis.

A total of 7 186 participants were included in this study, of whom 817（10.89%）had COPD. Multivariate logistic regression revealed that the prevalence of COPD in the highest quartile group was 40% lower compared to the lowest quartile group（OR=0.60, 95% CI:0.39-0.94, P=0.029）. Restricted cubic spline analysis confirmed a linear negative correlation between Log₂CALLY and COPD（P_non-linear=0.74）. No significant interactions were observed between age, gender, body mass index, smoking and drinking status, hypertension, diabetes, asthma, cardiovascular disease, and Log₂CALLY（P>0.05）. ROC curve analysis demonstrated that after full adjustment, the area under the curve（AUC）for Log₂CALLY in predicting COPD risk was 0.83.

The CALLY index is linearly negatively correlated with the risk of COPD, indicating its potential as a composite biomarker for early risk prediction of COPD.

The aim of this study is to comprehensively analyze the causal relationship between osteoporosis and cognitive impairment in China community using the combination of observational study and Mendelian randomization (MR).

Baseline data from the West China Health and Aging Cohort Study were used to explore phenotypic associations between L1-L4 bone mineral density (BMD), total hip BMD, femoral neck BMD, osteoporosis, osteopenia and cognitive function. A two-sample MR design was used to evaluate the causal association between BMD and Alzheimer’s disease (AD) using summary statistics from published genome-wide association studies, with single nucleotide polymorphisms (SNPs) significantly associated with BMD as instrumental variables and AD as the outcome variable.

The cross-sectional study showed that after adjusting for a series of confounding factors, osteoporosis was found to be negatively associated with Mini-Mental State Examination (MMSE) score compared with normal bone mass group (β=-0.84, 95% CI: -1.39~-0.29, P=0.003), indicating that the patients with osteoporosis had lower MMSE scores and worse cognitive function than those with normal bone mass, and the results were robust. The relationship between BMD at different sites and cognitive scores was different.Specifically, total hip BMD (β=1.58, 95% CI: 0.20-2.95, P=0.024) and L1-L4 BMD (β=1.17, 95% CI: 0.13-2.21, P=0.028) were positively correlated with MMSE score, while femoral neck BMD was not significantly correlated with MMSE score (β=1.34, 95% CI: -0.11-2.79, P=0.070). Two-sample MR analysis using 86 SNPs associated with heel BMD-T values in a Chinese population as instrumental variables showed no causal association between heel BMD-T and AD risk (IVW: OR=0.95, 95% CI: 0.84-1.07, P=0.417). The results of multiple sensitivity analyses were robust.

The evidence from this study suggests a positive association between osteoporosis and cognitive impairment in Chinese, but further studies are needed to confirm the causal relationship between the two traits. Based on the common risk factors and pathogenesis, it is beneficial in practice to strengthen the management of osteoporosis patients, especially those with low total hip BMD and L1-L4 BMD, to prevent the occurrence of cognitive impairment.

To investigate the relationship between dynamic changes in cholesterol, high-density lipoprotein, and glucose index (CHG) and new-onset cardiovascular disease (CVD) in the middle-aged and elderly Chinese population.

In this study, the China Health and Retirement Longitudinal Study (CHARLS) data were utilized. Both K-means clustering and cumulative measures were used to characterize the dynamic changes of CHG. Cox proportional hazards regression model was used to assess the association between CHG changes with CVD. The nonlinear relationship between cumulative CHG and CVD was further assessed using restricted cubic spline. Subgroup analyses were conducted to assess whether the association between CHG changes and CVD differed across participant subgroups.

A total of 4 710 participants were included, and a total of 785 (16.7%) developed CVD during the follow-up period. The patterns of changes in CHG over the two follow-up visits were categorized into three classes by K-means clustering: persistently low (class 1), persistently medium (class 2), and persistently high (class 3). The risk of CVD was significantly increased in class 2 (HR=1.38, 95% CI: 1.17-1.64) and class 3 (HR=1.44, 95% CI: 1.11-1.87), compared with class 1. With the first quartile as reference, the third and fourth quartile of cumulative CHG had an increased risk of CVD and stroke. RCS analysis revealed a significant positive linear relationship between cumulative CHG and the risk of CVD (P_nonlinearity=0.372) and stroke (P_nonlinearity=0.200). Subgroup analyses did not identify variables that significantly interacted with changes in CHG.

Dynamic changes in CHG are significantly associated with the risk of CVD. The simplicity and affordability of CHG make it particularly valuable for large-scale screening, providing a new reference index for the risk prediction of CVD in the future.

To investigate the longitudinal association between smoking cessation, secondhand smoke (SHS) exposure, and accelerated biological aging, and to further explore the heterogeneous associations of SHS exposure on biological aging among individuals who have quit smoking.

This study was based on data from the baseline and the first repeat assessment of the UK Biobank. Phenotypic Age (PA) was constructed using the validated PA algorithm, and PA acceleration was calculated. Multivariable linear regression models were used to perform change analysis to assess the associations between changes in smoking behavior and SHS exposure with changes in PA acceleration. Stratified analyses were conducted to identify potential effect heterogeneity.

A total of 11 189 eligible participants were included. Compared to never smokers, persistent smokers showed significantly greater PA acceleration (β=0.94, 95% CI: 0.22-1.65), whereas long-term quitters were associated with lower PA acceleration (β=0.30, 95% CI: 0.05-0.55). Compared with those consistently unexposed to SHS, individuals with increased SHS exposure exhibited greater PA acceleration(β=0.40, 95% CI: 0.01-0.79). Notably, among long-term quitters, frequent SHS exposure was associated with significantly greater biological aging (β=0.80, 95% CI: 0.02-1.57).

Smoking cessation can partially reverse the adverse effects of smoking on biological aging, although long-term damage may not be fully reversible. Moreover, SHS exposure significantly accelerates biological aging, with more pronounced effects observed among individuals who have quit smoking.

To investigate the chain mediating effect of family health and self-efficacy in the relationship between health literacy and self-management behaviors in older adults with chronic comorbidities.

Based on data from the 2021 Psychology and Behavior Investigation of Chinese Residents, 310 patients aged ≥60 years with chronic comorbidities were selected for the study. They were investigated using the Health Literacy Scale (HLS-SF12), Chronic Disease Self-Management Behavior Scale, Family Health Scale (FHS-SF), and New General Self-Efficacy Scale (NGSES). Partial correlation analysis was used to explore the correlation between health literacy, family health, self-efficacy, and self-management behaviors. The chain mediating effect was analyzed using Model 6 of the PROCESS 4.2 plugin in SPSS 27.0.

Family health and self-efficacy played a chain mediating role in the relationship between health literacy and self-management behaviors. The direct effect of health literacy on self-management behaviors was significant (β=0.446, 95% CI: 0.239-0.653), accounting for 66.87% of the total effect. The separate mediating effect of family health was significant (β=0.071, 95% CI: 0.004-0.154), accounting for 10.64% of the total effect. The separate mediating effect of self-efficacy was significant (β=0.117, 95% CI: 0.043-0.206), accounting for 17.54% of the total effect. The chain mediating effect of family health and self-efficacy was also significant (β=0.034, 95% CI: 0.010-0.067), accounting for 5.10% of the total effect.

Health literacy not only directly influences self-management behaviors of elderly patients with chronic comorbidities, but also indirectly through family health and self-efficacy as separate or chain-mediated pathways. It is recommended to promote the self-management behaviors of elderly patients with chronic comorbidities in a multidimensional way by enhancing health literacy, improving the family health environment, and increasing self-efficacy.

To explore the impact of arthritis, depression and their interaction on the disability risk in Chinese elderly.

Using data from the China Health and Retirement Longitudinal Study (CHARLS), eligible participants were enrolled, and 2015 baseline data and incident disability during the 5-year follow-up were collected. Multivariable cox regression models were established to analyze the effects of arthritis and depression on disability risk among Chinese elderly and to assess, on the multiplicative and additive scales, the impact of their interaction on disability risk.

A total of 7 875 participants were included, and 586 new cases of disability occurred during the 5-year follow-up period, with an incidence rate of 7.4%.The multivariable cox regression analysis showed that arthritis (HR=1.92, 95% CI: 1.63-2.26) and depression (HR=2.65, 95% CI: 2.22-3.16) were both risk factors for disability. The interaction analysis showed that there was no multiplicative interaction between arthritis and depression on the disability risk (P>0.05), but there was an additive interaction effect (P<0.05), with the relative excess risk due to interaction (RERI) of 1.08 (95% CI: 0.32-1.85), the attributable proportion of interaction (AP) of 0.26 (95% CI: 0.09-0.42), and the synergy index (SI) of 1.51 (95% CI: 1.09-2.08).

Both arthritis and depression are risk factors for disability in Chinese elderly, and there is an additive interaction between them. We should pay more attention to elderly people with comorbid arthritis and depression and take active measures to reduce the disability risk in Chinese elderly.