Acute ischemic stroke (AIS) is one of the most time-sensitive diseases, which requires prompt medical services. Delayed treatment and resource scarcity during the coronavirus disease 2019 (COVID-19) pandemic may affect the quality of AIS care and increase the risk of adverse outcomes for patients. The study used data from China’s National Medical Quality Database spanning from January 1, 2019, to May 31, 2022. Quality of care was gauged utilizing 15 indicators, featuring 12 quality indicators (QIs) and 3 composite indicators. We used mixed-effects and interrupted time series models to compare care quality pre- and post-pandemic. Utilization rates of 15 indicators were compared across different age and gender groups to evaluate healthcare disparities. Multilevel Logistic regression and structural equation modeling assessed the impact of the 3 composite indicators on outcome (in-hospital mortality or 30-day readmission) during the pandemic. Utilization rates for 10 of 12 QIs and three composite indicators declined rapidly after the COVID-19 outbreak, but later improved. The most significant drop was observed for rehabilitation (QI12), with a decrease of 50.06

BACKGROUND:Acute coronary syndrome (ACS) is a common and serious cardiovascular disease with high morbidity and mortality. Timely diagnosis and treatment can help reduce the mortality of patients, but the arrival of the holidays may affect the quality of ACS treatment. OBJECTIVES:To analyse the trends in the treatment quality of ACS care and the holiday season effect. METHODS:Joinpoint regression was performed to measure the overall trend in adherence rate of QIs from 1 January 2016, to 30 June 2020. The trends in emergency care quality over time was quantised by average monthly percentage change (AMPC). Interrupted time series were used to analyse the holiday season effect of the Spring Festival. RESULTS:The completion rates of five QIs (QI2, QI4, QI5, QI7, QI8) showed a continuous upward trend between 1 January 2016 and 30 June 2020, with QI4 showing the fastest upward trend between month 35 and month 38 (AMPC = 24.31%, p < 0.05). The completion rates of QI3, QI6 and QI9 showed an increasing trend at the beginning of the study and a decreasing trend at the later stages. Results of ITS model showed that there were immediate changes in rates of nine QIs before and after Chinese Spring Festival, with the largest decrease was in QI5 in phase 2 (level change = -13.94%, p < 0.05) and the highest increase was in QI4 in phase 4 (level change = 27.71%, p < 0.05). Excluding the effects of the holiday, completion rates continue to increase across the eight QIs. CONCLUSIONS:Quality of care in ACS declined around the Spring Festival. Excluding the effect of holidays, the quality of ACS disease care showed an improving trend, especially in integrated care and multidisciplinary collaboration.

Existing studies in developing countries on the impact of chest pain center (CPC) accreditation on treatment quality have limited ability to demonstrate causal relationships. This retrospective study aims to utilize the data from national-level database and explore the impact of chest pain center certification on the treatment quality of ST-segment elevation myocardial infarction (STEMI) patients through a more appropriate method. At the hospital level, taking timely reperfusion and in-hospital mortality as outcomes, the impact was evaluated using the Counterfactual Synthetic Difference-in-Differences (CS-DID) method, a statistical technique that allows for the estimation of causal effects by comparing the differences over time between treated and non-treated groups. The results showed that CPC accreditation improved timely reperfusion of STEMI. Once a CPC was certified, without considering covariates, the timely reperfusion rate increased on average by 5.4%, the 90-min PCI rate by 7.1%, and the 30-min thrombolysis rate by 2.0% in comparison with non-accredited hospitals, and this effect shows a downward trend over time and varies between different regions. We found no evidence to confirm that CPC accreditation decreases in-hospital mortality in patients with STEMI. CPC accreditation in China has improved the timeliness of reperfusion therapy for STEMI patients. CPC accreditation and re-accreditation are crucial to maintaining high-quality care for STEMI patients.

Background:The progression of knee osteoarthritis is mainly characterized by the reduction in joint space width (JSW). The goal of this study was to build a knee joint space segmentation model through deep learning (DL) methods and develop a model for automatically measuring JSW. Furthermore, we predicted JSW changes in the sixth year based on regression models. Methods:The data for this study was sourced from the Osteoarthritis Initiative database. We filtered knee X-ray images from 1,947 participants and tested six neural networks for segmentation to build an automatic JSW measurement model. Subsequently, we combined the clinical data with the JSW measurement results to predict the sixth-year knee JSW using six different regression models. Results:The segmentation results showed that TransUNet performed the best, with an overall Dice coefficient of 0.889. The intraclass correlation coefficient (ICC) between manually measured and TransUNet's automatically measured JSW reached 0.927 (P<0.01). Among the regression models, eXtreme Gradient Boosting (XGBoost) demonstrated the best predictive performance, with a mean absolute error (MAE) of 0.48 and an ICC of 0.887 (P<0.01). To better align with clinical practice, we reduced the prediction model to utilize only 2 years of JSW images. The results showed that using the 0- and 12-month X-ray images still achieved high accuracy, with an MAE of 0.585 (P<0.05) and an ICC of 0.805 (P<0.01). Conclusions:We developed a novel JSW measurement model that significantly improves accuracy compared to previous methods and identified the best prediction model by combining TransUNet and XGBoost. Additionally, in our built model, predicting the 72-month JSW using only 2 years of knee X-ray images and several clinical features achieved high accuracy.

High-dimensional proteomics data present significant challenges in biomarker discovery due to technical noise, feature redundancy, and multicollinearity. Current feature selection methods, including filter, wrapper, and embedded approaches, struggle with stability, sparsity, and computational efficiency. To address these limitations, we propose Soft-Thresholded Compressed Sensing (ST-CS), a hybrid framework integrating 1-bit compressed sensing with K-Medoids clustering. Unlike conventional methods relying on manual thresholds, ST-CS automates feature selection by dynamically partitioning coefficient magnitudes into discriminative biomarkers and noise. Evaluations on simulated and real-world proteomic datasets demonstrated ST-CS’s superiority in feature selection capability and classification performance. In simulations, ST-CS achieved feature selection robustness with balanced sensitivity (> 80

BACKGROUND:Stroke remains a leading global cause of death, with treatment timeliness critically determining outcomes. Although the time-efficacy relationship in stroke care is well established, the interplay interaction between treatment delays, care quality and clinical outcomes remains poorly characterised, particularly across different healthcare settings. METHODS:This nationwide study analysed data from 2 875 119 acute ischaemic stroke (AIS) hospitalisations (2020-2024). After stratifying patients by treatment delay quartiles (Q1-Q4), we performed propensity score matching to balance 24 baseline covariates. To evaluate the detrimental effects of treatment delay on therapeutic benefits, we used logistic regression and doubly robust causal modelling across delay groups. Patients whose overall delay fell within the fourth quartile and received low-quality care were identified as the high-risk group. Multivariable logistic regression was used to identify independent risk factors. RESULTS:Low-quality care correlated with longer delays (overall: 1038 vs 981 min, p<0.0001). High-quality care achieved the greatest mortality reduction in Q1 (average treatment effect (ATE) 0.0036, 95% CI 0.0032 to 0.0041) compared with Q4 (ATE 0.0014, 95% CI 0.0012 to 0.0017). Thrombolysis delays had the strongest impact on mortality (Q1 ATE 0.0155, 95% CI 0.0087 to 0.0222; Q4 ATE 0.0068, 95% CI 0.0031 to 0.0106). High-risk subgroups for delayed, low-quality care included: Northwest residents (OR 1.5759, 95% CI 1.5613 to 1.5905), minor stroke (OR 1.8402, 95% CI 1.8302 to 1.8503), self-transport patients (OR 1.1392, 95% CI 1.1340 to 1.1443), and those with comorbidities (renal failure: OR 1.0948, 95% CI 1.0825 to 1.1073; asthma: OR 1.0861, 95% CI 1.0646 to 1.1080) (all p<0.0001). CONCLUSIONS:The benefits of high-quality care in reducing mortality risk were significantly diminished by delays in hospital admission, examination and thrombolysis. The timeliness and quality of AIS care are influenced by geographic location, admission National Institutes of Health Stroke Scale scores and comorbidity profiles. The highest priority populations for delay reduction and quality improvement were patients who did not use emergency medical services and those with multiple comorbidities.

BackgroundKnee cartilage is the most crucial structure in the knee, and the reduction of cartilage thickness is a significant factor in the occurrence and development of osteoarthritis. Measuring cartilage thickness allows for a more accurate assessment of cartilage wear, but this process is relatively time-consuming. Our objectives encompass using various DL methods to segment knee cartilage from MRIs taken with different equipment and parameters, building a DL-based model for measuring and grading knee cartilage, and establishing a standardized database of knee cartilage thickness.MethodsIn this retrospective study, we selected a mixed knee MRI dataset consisting of 700 cases from four datasets with varying cartilage thickness. We employed four convolutional neural networks—UNet, UNet++, ResUNet, and TransUNet—to train and segment the mixed dataset, leveraging an extensive array of labeled data for effective supervised learning. Subsequently, we measured and graded the thickness of knee cartilage in 12 regions. Finally, a standard knee cartilage thickness dataset was established using 291 cases with ages ranging from 20 to 45 years and a Kellgren–Lawrence grading of 0.ResultsThe validation results of network segmentation showed that TransUNet performed the best in the mixed dataset, with an overall dice similarity coefficient of 0.813 and an Intersection over Union of 0.692. The model’s mean absolute percentage error for automatic measurement and grading after segmentation was 0.831. The experiment also yielded standard knee cartilage thickness, with an average thickness of 1.98 mm for the femoral cartilage and 2.14 mm for the tibial cartilage.ConclusionBy selecting the best knee cartilage segmentation network, we built a model with a stronger generalization ability to automatically segment, measure, and grade cartilage thickness. This model can assist surgeons in more accurately and efficiently diagnosing changes in patients’ cartilage thickness.

Objectives The quality of care for patients may be partly determined by the time they are admitted to the hospital. This study was conducted to explore the effect of admission time and describe the pattern and magnitude of weekly variation in the quality of patient care. Study design A retrospective observational study. Methods Data were collected from the Medical Care Quality Management and Control System for Specific (Single) Diseases in China. A total of 238,122 patients treated for acute ischemic stroke between January 2015 and December 2017 were included. The primary outcomes were completion of the ten process indicators and in-hospital death. Results The quality of in-hospital care varied according to hospital arrival time. We identified several patterns of variation across the days of the week. In the first pattern, the quality of four indicators, such as stroke physicians within 15 min, was lowest for arrivals between 08:00 and 11:59, increased throughout the day, and peaked for arrivals between 20:00 and 23:59 or 00:00 and 03:59. In the second pattern, the quality of four indicators, such as the application of antiplatelet therapy within 48 h, was not significantly different between days and weeks. There was no difference in in-hospital mortality between the different admission times. Conclusions The effect of admission time on the quality of in-hospital care of patients with acute ischemic stroke showed several diurnal patterns. Detecting the times when quality is relatively low may lead to quality improvements in health care. Quality improvement should also focus on reducing diurnal temporal variation.

Background: Reporting the results of quality indicators can narrow the gap in the quality of care between hospitals. While most studies rely on outcome indicators , they may not accurately measure the quality of care. Process indicators are not only strongly associated with treatment outcomes, but are also more sensitive to whether patients are treated accurately, enabling timely intervention. In this study, we aimed to investigate whether process indicators can provide a more reasonable assessment of evaluating hospital quality of care compared to outcome indicators. Methods: Data were sourced from the Specific Disease Medical Service Quality Management and Control System in China. A total of 113,942 patients with breast cancer treated in 298 hospitals between January 2019 and April 2023 were included in this study. The rankability of 11 process indicators was calculated and used as a weight to create a new composite indicator. The composite indicators and outcome measures were compared using the O/E ratio categories. Finally, in order to determine the impact of different years on the results, a sensitivity analysis was conducted using bootstrap sampling. Results: The rankability (ρ) values of the eleven process indicators showed significant differences, with the highest ρ value for preoperative cytological or histological examination before surgery (0.919). The ρ value for the outcome indicator was 0.011. The rankability-weighting method yielded a comprehensive score (ρ = 0.883). The comparison with categorical results of the outcome indicator has different performance classifications for 113 hospitals (37.92%) for composite 1 scores and 140 (46.98%) for preoperative cytological or histological examination before surgery. Conclusion: Process indicators are more appropriate than outcome indicators for assessing the quality of hospital breast cancer care, which may contribute to quality improvement efforts.

Major depressive disorder (MDD) is one of the leading causes of disability worldwide. Comprehensive description of the global burden of MDD and its attributable risk factors is essential for policymaking but currently lacking. In this study, we aim to estimate the burden of MDD in terms of incidence, prevalence, and years lived with disability (YLDs), along with its attributable risk factors at global, regional, and rational level between 1990 and 2019, using data from the Global Burden of Diseases, Injuries, and Risk Factors Study 2019. Data analysis was completed on July 1, 2023. In 2019, 274.80 million (95 % uncertainty interval [UI], 241.28 to 312.77) new cases of MDD were identified globally, with an increase of 59 % from 1990. A total of 37.20 million (25.65 to 51.22) YLDs were attributable to MDD, accounting for the largest proportion of mental disorder YLDs (29.7 %). Countries in the low sociodemographic index quantile exhibited the highest age-standardized incidence rate of MDD, with Uganda (7836.2, per 100,000 person-years, 6713.7 to 9181.1) and Palestine (7687.7, 6546.1 to 9023.9) reporting the highest rates among them. The United States had the highest increase in age-standardized rates, with an average annual percent change of 0.99. Females had 1.6 times higher age-standardised rates than males, ranging from 1.2 (Oceania) to 2.2 (tropical Latin America) times across 21 regions. Globally, the proportions of YLDs due to MDD attributable to bullying victimization, childhood sexual abuse, and intimate partner violence were 4.86 %, 5.46 %, and 8.43 % in 2019, respectively. The heavy burden of MDD serves as a stark reminder that a coordinated response from governments and health communities is urgently needed to scale up mental health services and implement effective interventions, particularly in low-income countries.

Background Osteoarthritis (OA) is a degenerative disease requiring additional research. This study compared gene expression and immune infiltration between lesioned and preserved subchondral bone. The results were validated using multiple tissue datasets and experiments. Methods Differentially expressed genes (DEGs) between the lesioned and preserved tibial plateaus of OA patients were identified in the GSE51588 dataset. Moreover, functional annotation and protein–protein interaction (PPI) network analyses were performed on the lesioned and preserved sides to explore potential therapeutic targets in OA subchondral bones. In addition, multiple tissues were used to screen coexpressed genes, and the expression levels of identified candidate DEGs in OA were measured by quantitative real-time polymerase chain reaction. Finally, an immune infiltration analysis was conducted. Results A total of 1,010 DEGs were identified, 423 upregulated and 587 downregulated. The biological process (BP) terms enriched in the upregulated genes included “skeletal system development”, “sister chromatid cohesion”, and “ossification”. Pathways were enriched in “Wnt signaling pathway” and “proteoglycans in cancer”. The BP terms enriched in the downregulated genes included “inflammatory response”, “xenobiotic metabolic process”, and “positive regulation of inflammatory response”. The enriched pathways included “neuroactive ligand–receptor interaction” and “AMP-activated protein kinase signaling”. JUN, tumor necrosis factor α, and interleukin-1β were the hub genes in the PPI network. Collagen XI A1 and leucine-rich repeat-containing 15 were screened from multiple datasets and experimentally validated. Immune infiltration analyses showed fewer infiltrating adipocytes and endothelial cells in the lesioned versus preserved samples. Conclusion Our findings provide valuable information for future studies on the pathogenic mechanism of OA and potential therapeutic and diagnostic targets.

目的 构建适用于含隐变量混合数据类型的因果推断方法,通过模拟研究评估方法的效果,为观察性数据的因果结构研究提供依据.方法 基于约束的FCI-stable结构学习方法,结合条件高斯独立性检验,建立适用于含隐变量混合数据类型的因果推断算法框架.利用Tetrad 6.9.0生成含隐变量的混合类型模拟数据,模拟研究设置不同的离散型数据占比、隐变量占比、网络节点数量及样本量;将FCIcg方法与基于约束的FCIchi-square、FCIdg、FCIFfisher-z、FCIkci及FCImrcit五种方法进行比较,采用F1 Score和马修斯相关系数(matthews correlation coefficient,MCC)对因果结构识别效果进行评估.结果 本研究建立了适用于含隐变量混合数据类型的因果推断方法FCIcg,该方法可用于从观察性数据中学习变量间的因果结构.模拟结果表明,在不同离散型数据占比下,除FCIchi-square和FCIFfisher-z外,其他方法估计效果较稳定,FCIcg和FCIkci的估计效果优于FCIdg和FCImrcit;在有无隐变量和隐变量占比随之增加的情况下,除FCImrcit外其他方法估计效果的变化较平缓,FCIcg和FCIFkci的估计效果好于其他方法;在不同网络大小的情况下,除FCImrcit外其他方法评估效果较稳定;随着样本数量的增加,FCIcg和FCIFfisher-z估计效果稳步提升;当样本数量为2000,网络节点数量为14时,FCIcg的平均运行时间为(0.580±0.301)秒.结论 本文构建的因果推断方法针对含有隐变量的混合数据类型具有良好的效果,算法运行速度快;本方法可作为医学领域观察性数据因果结构识别的推荐方法.

目的 筛选铁死亡相关长链非编码RNA(ferLncRNA)构建风险评分模型,并进行肺腺癌患者的生存预测.方法 通过相关性分析、癌组织-癌旁组织差异分析,获得差异表达的铁死亡相关LncRNA(DEferlncRNA),使用配对算法获得DEferlncRNA pairs.利用LASSO Cox筛选与生存相关DEferlncRNA pairs,多因素Cox回归分析并构建风险评分模型;计算ROC曲线的AUC值评估模型.利用单因素和多因素Cox回归分析风险评分是否具有独立预后价值;分析生存时间、临床病理特征和化疗疗效等在高、低风险组之间的差别.结果 相关性分析获取 512 个ferlncRNA,差异分析获得 53 个DEferlncRNA,配对算法获得1243 个DEferlncRNA pairs.LASSO Cox分析获得40 个与生存相关DEferlncRNA pairs,多因素分析得到 17 个DEferlncRNA pairs构建风险评分模型;模型的AUC为 0.776.单因素和多因素分析提示风险评分可作为肺腺癌(lung adenocarcinoma,LUAD)的独立预后因素.患者化疗药物敏感性、TNM分期、临床分期、生存时间和生存状态在高、低风险组之间差异有统计学意义.结论 利用DEferlncRNA构建的肺腺癌风险评分模型,可为预测肺腺癌患者预后提供依据.此外,低风险评分患者对化疗药物敏感,将为肺腺癌患者临床治疗提供参考.

Background: Understanding trend characteristics of depression among cancer survivors is essential for healthcare policies and planning. This study estimates longitudinal trends in the prevalence and treatment of depression among adults in the United States with and without cancer.Methods: This cross-sectional study focused on adults aged 20 years or older based on nationally representative data from the National Health and Nutrition Examination Surveys 2005-2020. Weighted logistic regression model was established to assess association between depression and cancer status after adjusting various covariates potentially related to depression.Results: Among the 37,283 participants (weighted mean age, 47.5; women, 50.9 %), 3648 (9.8 %) were diagnosed with cancer and 3343 (9.0 %) were screened positive for depression. The age-standardized prevalence of depression showed a U-shaped trend in cancer survivors, decreasing from 11.8 % (95 % confidence interval, 8.4 %-15.2 %) in 2005-2008 to 8.3 % (5.6 %-11.0 %) in 2013-2016, then increasing to 11.7 % (6.3 %-17.2 %) in 2017-2020. These trends varied by population subgroup. Among depressive patients with cancer, antidepressant use increased from 38.6 % (28.7 %-48.5 %) in 2005-2008 to 62.9 % (40.6 %-85.2 %) in 2017-2020, whereas mental health consultation increased slightly. Limitations: Using a screening questionnaire instead of diagnostic criteria to identify depression; small sample size of patients with cancer; and cross-sectional analysis without prospective outcomes.Conclusions: From 2005 to 2020, the depression disease burden in patients with cancer eased in 2009-2015, but deteriorated recently. A healthy lifestyle and reasonable treatment for depression, based on an objective examination of depression characteristics, would improve long-term cancer outcomes and quality of life.

目的 分析急性心肌梗死(acute myocardial infarction,AMI)患者治疗过程质量评价指标对患者再入院的影响,为提高治疗过程质量、改善患者结局提供依据.方法 收集某医院2014年12月-2016年12月期间急性心肌梗死患者;选取到达医院10min内心电图检查、冠脉造影等7个治疗过程指标,计算指标使用率;利用Probit回归与工具变量Probit模型分析治疗过程评价指标对再入院的影响.结果 纳入4317例急性心肌梗死患者,其中245例出院后再次住院,再入院率为5.68％.治疗过程指标为:到达医院10min内心电图检查、冠脉造影检查、入院24h内β受体阻滞剂、住院期间氯吡格雷的使用、到达医院30min内溶栓治疗、入院90min内PCI和入院24h内阿司匹林,其使用率分别为99.68％、82.99％、44.49％、97.29％、0.07％、63.99％和98.34％.Probit回归分析显示,接受冠脉造影检查(OR=0.778,95％CI=0.661～0.917)和入院90min内PCI治疗(OR=0.808,95％CI=0.689～0.950)的患者再入院风险较未接受患者降低;住院期间氯吡格雷的使用与再入院成正比且差异有统计学意义(OR=1.354,95％CI=1.023～1.428).工具变量Probit模型分析显示,治疗过程指标中到达医院30min内溶栓治疗、入院24h内β受体阻滞剂、住院期间氯吡格雷的使用、入院24h内阿司匹林具有内生性,接受相对未接受4种药物治疗的患者再入院风险降低;OR值分别为:OR=0.4(95％CI:0.163～0.984)、OR=0.103(95％CI:0.011～0.939)、OR=0.729(95％CI:0.536～0.992)、OR=0.286(95％CI:0.084～0.977).结论 在急性心肌梗死患者的治疗过程中,可以通过提高冠脉造影、入院90min内PCI、到达医院30min内溶栓治疗、入院24h内β受体阻滞剂、住院期间氯吡格雷、入院24h内阿司匹林的使用,从而改善治疗过程质量、降低再入院风险.

目的 筛选鸡肉不耐受患者与健康人之间的差异蛋白,富集分析识别差异蛋白的相关通路,为探讨鸡肉不耐受发生机制提供依据.方法 收集鸡肉不耐受患者及健康对照的血清样本,利用DIA技术平台对样本进行蛋白组学检测;PLS-DA结合PPI分析筛选具有相互作用的蛋白质,Cytoscape获得鸡肉不耐受患者与对照组间的差异蛋白;GO和KEGG富集分析获得差异蛋白的生物学通路与代谢通路.结果 血清样本的蛋白表达数据检测结果,共获得 2210 种蛋白质.PLS-DA分析筛选出和鸡肉不耐受相关的蛋白质 786 个,结合蛋白互作分析获得PPI网络,聚类后得到两个主要的模块:cluster1和cluster2;cluster1包括18个蛋白质,cluster 2包括12个蛋白质.两模块蛋白质进行重要性排序,筛选出鸡肉不耐受患者与健康对照间的差异蛋白为:ORM1、FGB、SERPINF2、HRG、A2M、ORM2和QSOX1.前五个为上调蛋白质,后两个为下调蛋白质.cluster1评分为10.12,GO富集分析发现差异蛋白主要参与血小板脱颗粒、肽链内切酶活性的负调节、炎症反应等生物学通路,KEGG分析富集到补体和凝血级联通路;cluster 2评分6.00,共包括12个蛋白质,GO富集分析发现差异蛋白与急性期反应、血小板脱颗粒、固有免疫应答等生物学通路有关,KEGG分析富集到补体和凝血级联通路、阮病毒病和系统性红斑狼疮的疾病通路.结论 本研究筛选出7个与IgG介导的鸡肉不耐受相关的差异蛋白,其参与的生物学和代谢通路与机体免疫、炎症过程有关,在免疫、炎症发生发展中可能起重要作用,将为探讨鸡肉不耐受发生机制提供依据.

Background The Ministry of Health of China conducted a study targeting in single-disease quality control in 2009, aimed to strengthen quality management and improve health care services. This study retrospectively investigated the trends of quality indicators for six monitored diseases 2011-2017 to evaluate the improvement of care quality for the first batch of single-disease. Methods We extracted data from the National Specific (Single) Disease Monitoring System for 2011-2017. We focused on six conditions: acute myocardial infarc-tion, heart failure, community-acquired pneumonia, coronary artery bypass graft, hip / knee replacement, and acute ischemic stroke. A total of 56 quality indicators (QIs) were adopted to monitor the quality change and determine the trends in care quality. We also calculated the hospital process composite performance (HPCP) using a denominator-based weighting method for each hospital per year. The es-timated annual percentage changes (EAPC) 2011-2017 were calculated at nation-al and regional levels.Results The results showed that use of four QIs had significant downward trends, whereas 25 QIs (including reversed indicators) showed significant upward trends from 2011 to 2017. The greatest improvement was observed in CAP-4 (antibiotic treatment within four hours after admission to the hospital for critical pneumonia) in the central region (EAPC = 48.36, 95% CI = 15.92-89.87); while the largest de-crease appeared in AIS-1 (thrombolytic therapy within 4.5 hours of symptom onset) in the western region (EAPC =-13.44, 95% CI =-24.98,-0.11). An increased HPCP was observed in four diseases nationwide, but not for acute myocardial infarction and heart failure. However, there were significant differences across regions in the process of care and outcomes, with the performance of Eastern and Western regions showing remarkable advantages compared with the Central region.Conclusions We provide evidence for major advancement in care quality in Chi-na nationwide. However, the improvement of care in China was unbalanced geographically and should be carefully considered. Future challenges include expanding the coverage of quality monitoring, greater delivery efficiency, and re-gion-balanced health care.

Background: Adherence to evidence-based hospital stroke care is variable and may change over time. It is important to determine which process measures are associated with variation in outcome. In a large dataset, we analyzed the association between process and outcome and the fluctuations of indicators over time, and identified quality indicators (QIs) that should be prioritized for improving the quality of stroke care. Methods: We analyzed data from 123,259 patients diagnosed with acute ischemic stroke (AIS) who were treated at 109 large tertiary hospitals in China between January 2011 and May 2017. In total, 12 stroke treatment indicators were selected to calculate the hospital process composite performance (HPCP). Hospitals were divided into subgroups according to the time trend of HPCP estimated by the Group-Based Model. We analyzed the influence of hospital subgroups on the patient outcomes using a multi-level model and explored the QIs that led to variation. Results: The HPCP trends for stroke indicators of 109 hospitals over 7 years were divided into two groups (Group 1, low-HPCP; Group 2, high-HPCP). After adjusting for patient age, medical insurance, comorbidities, patterns of admission, and NIHSS-scores, patients in the high-HPCP group presented higher rate of independence and longer length of stay compared to the low-HPCP group. The multi-level model showed that there was a statistically significant difference in the utilization rate between the two groups, with most marked differences seen in emergency assessment and function evaluation indicators. Conclusion: Variation in the quality of stroke care exists across hospitals, and better adherence to guideline-based care is associated with improved outcomes. We found that QIs related to emergency examination and functional assessment were the main factors differing between good and poor adherers to stroke indicators, suggesting that quality improvement in stroke care could prioritize these QIs.