Invasive coronary angiography imposes risks and high medical costs. Therefore, accurate, reliable, non-invasive, and cost-effective methods for diagnosing coronary stenosis are required. We designed a machine learning-based risk-prediction system as an accurate, noninvasive, and cost-effective alternative method for evaluating suspected coronary heart disease (CHD) patients. Electronic medical record data were collected from suspected CHD patients undergoing coronary angiography between May 1, 2017, and December 31, 2019. Multi-Class XGBoost, LightGBM, Random Forest, NGBoost, logistic models and MLP were constructed to identify patients with normal coronary arteries (class 0: no coronary artery stenosis), minimum coronary artery stenosis (class 1: 0 < stenosis <50%), and CHD (class 2: stenosis ≥50%). Model stability was verified externally. A risk-assessment and management system was established for patient-specific intervention guidance. Of 1577 suspected CHD patients, 81 (5.14%) had normal coronary arteries. The XGBoost model demonstrated the best overall classification performance (micro-average receiver operating characteristic [ROC] curve: 0.92, macro-average ROC curve: 0.89, class 0 ROC curve: 0.88, class 1 ROC curve: 0.90, class 2 ROC curve: 0.89), with good external verification. In class-specific classification, the XGBoost model yielded F1 values of 0.636, 0.850, and 0.858, for Classes 0, 1, and 2, respectively. The visualization system allowed disease diagnosis and probability estimation, and identified the intervention focus for individual patients. Thus, the system distinguished coronary artery stenosis well in suspected CHD patients. Personalized probability curves provide individualized intervention guidance. This may reduce the number of invasive inspections in negative patients, while facilitating decision-making regarding appropriate medical intervention, improving patient prognosis.

Objective: Shared decision‐making (SDM) as a multicollaborative approach is vital for facilitating patient‐centred care. Considering the limited clinical practice, we attempted to synthesize the motivations and resistances, and investigate their mutual relationships for advancing the implementation of SDM.Methods: A comprehensive systematic review using Preferred Reporting Items for Systematic Review and Meta‐Analysis guidelines was performed. ‘Shared decision making’ was searched as the mesh term through PubMed, Web of Science and EBSCO from 2000 to 2021, and the quality of literature was appraised using the QualSyst Tool. Motivations and resistances were categorized based on content analysis and the ‘structure–process–outcome’ model.Results: From 8319 potential citations, 105 were included, comprising 53 qualitative studies (the average quality score is 0.92) and 52 quantitative studies (the average quality score is 0.95). A total of 42 categories of factors were identified into 11 themes and further grouped into three dimensions: structure, process and outcome. The structure dimension comprised six themes (71.43%), the process dimension contained four themes (11.01%) and the outcome dimension covered only one theme. Across all categories, decision‐making time and patients' decision preparedness in the process dimension were the most reported, followed by physicians' communication skills and health care environment in the structure dimension. Analysis of implementation of SDM among various types of diseases showed that more influencing factors were extracted from chronic diseases and unspecified disease decisions.Conclusions: The major determinants for the implementation of SDM are focused on the structural dimension, which challenges the health systems of both developed and low‐ and middle‐income countries. Furthermore, we consider it important to understand more about the interactions among the factors to take integrated measures to address the problems and to ensure the effectiveness of implementing SDM.Patient or Public Contribution: Patients, healthcare professionals and other stakeholders articulated their perspectives on the implementation of SDM actively, and these were adopted and analysed in this study. However, the above‐mentioned individuals were not directly involved in the process of this study. Protocol was registered on PROSPERO (CRD42021259309).

The aims of this study were to investigate and compare patient satisfaction with outpatient care in public secondary and tertiary hospitals in China and to explore the factors affecting patient satisfaction for improving the quality of outpatient care in public hospitals. This cross-sectional study comprised a sample survey of 11 097 adults in 31 provincial cities in China from February to April 2018. A pretested structured questionnaire was used to collect outpatient experience data through a computer-assisted telephone interviewing system. Patient satisfaction was assessed using nine questions answered on a 4-point Likert scale. Multivariate regression models were employed to examine the relationships of patient satisfaction with outpatient services and healthcare provider level and to identify the factors associated with satisfaction. Patient’s overall satisfaction score with outpatient care was 27.3 (SD = 3.8), with lower scores observed in tertiary hospitals than in secondary hospitals (27.3 vs. 27.6, P < 0.05). The domain with the highest satisfaction was ‘consulting environment’, and the domain with the lowest satisfaction was ‘patient waiting time in the hospital’. Patients who went to tertiary hospitals reported lower satisfaction in ‘patient waiting time in the hospital’, ‘medical expenses’, ‘patient length of treatment time’ and ‘attitudes of other health workers’ than patients who went to secondary hospitals (P < 0.05). In secondary hospitals, no significant difference in patient satisfaction was observed between different sociodemographic categories (P > 0.05). In tertiary hospitals, female and single respondents were more likely to have higher satisfaction (P < 0.05), whereas respondents with high school or junior college degrees were more likely to have lower satisfaction (P < 0.05). The aforementioned results suggested that tertiary hospitals face larger challenges in patient satisfaction with outpatient care than secondary hospitals. Measures must be adopted to improve patient satisfaction with outpatient care in future healthcare reforms. Patient waiting time, medical expenses and treatment duration especially require improvements in tertiary hospitals.

Background: Blood glucose monitoring is essential in diabetic care and management. Monitoring using glucometers in home and in laboratories by professionals in certain health institutes were the common methods of blood glucose monitoring in clinical practice. This study aimed to characterize the profiles of blood glucose monitoring in the view of the discrepancy in methods and frequency conducted by the patients with type 2 diabetes mellitus (T2DM) in China, and to explore factors influencing the profiles. Methods: A cross-sectional, community-based study was conducted in Shandong province, China, with a multi-stage stratified sampling. A total of 2166 T2DM patients completed the structured questionnaires about the real-world status of blood glucose monitoring and other questions composed of demographic and clinical characteristic as well as the diabetes-related cognitive scales. Latent profile analysis (LPA) was used to identify the underlying profiles of blood glucose monitoring based on self-reported frequency of blood glucose monitoring through different methods. Univariate and multivariate logistic regression were used to analyze the characteristics of the profiles and to explore the factors associated with it. Results: Among the 2166 participants, the mean frequency of blood glucose monitoring was 2.77 times (standard deviation: 7.67) per month. LPA indicated that five-class model was the best solution for classifying the latent groups of blood glucose monitoring: Class 1 “Low frequency in all”, Class 2 “High frequency in hospitals”, Class 3 “High frequency in primary health institutes”, Class 4 “High frequency in pharmacies”, and Class 5 “High frequency in self-monitoring”. The proportions of the patients in class 1, class 2, class 3, class 4, and class 5 were 88.1% (n=1909), 1.3% (n=28), 3.1% (n=67), 6.1% (n=133) and 1.3% (n=29), respectively. Multivariate logistic regression showed that participants who had higher income (OR: 1.58, 95% CI: 1.04∼ 2.41, p< 0.05), had diabetes complication(s) (OR=1.37, 95% CI: 1.03∼ 1.02, p=0.03) and had a good knowledge of blood glucose control (OR=1.59, 95% CI: 1.17∼ 2.16, p< 0.01) were more likely to have high frequency of blood glucose monitoring (in class 2, 3, 4, 5), and the rural patients were less likely to had high frequency of blood glucose monitoring (OR=0.47, 95% CI: 0.35∼ 0.63, p< 0.01). Conclusion: Low frequency dominates the characteristics of the profiles of blood glucose monitoring among T2DM patients in China, though distinct blood glucose monitoring groups can be identified by LPA. Educational and financial supports were recommended to increase the frequency of blood glucose monitoring in patients with T2DM, focusing on the patients with low socioeconomic status.

Background: The integration of disparate emergency resources and the improvement of emergency response teamwork are the underlying trends and shared requirements for building resilience in an era of multiple global public health crises.Objective: This study investigated the emergency response with emergency collaboration networks of each functional module and the overall Joint Epidemic Prevention and Control Mechanism (JPCM) network in China’s COVID outbreak prevention and control.Methods: The study employed a scholarly framework of “the integration of JPCM coordination and emergency collaborative modularization” to explore the attributes of JPCM using social network analysis. The data were obtained from administrative records from JPCM’s official website, spanning January 2020 to December 2022.Results: The study examined the JPCM coordination and found several functional working modules of JPCM, such as Interrupt Spread, Manage Supply, Medical Rescue, Restore Work and Production, and Implement Responsibility modules. The network structure indicators showed that the Manage Supply module had the most extensive network connectivity, the shortest communication distance, and the most consistent collaboration. The E-I index of the overall JPCM network and the Manage Supply network were − 0.192 and − 0.452, respectively (at p < 0.001 and p < 0.05), indicating more internal relationships than external relationships. The E-I index of the Medical Rescue and Implement Responsibility collaboration networks were 0.122 and 0.147, respectively (at p < 0.001 and p < 0.05), indicating more external relationships than internal relationships. The QAP regression analysis showed that the most vital driver on the overall JPCM network was the Interrupt Spread module, followed by the Implement Responsibility and Medical Rescue modules.Discussion: The Interrupt Spread module initiated emergency coordination with most departments and agencies. The Manage Supply module ensured the flow of medical supplies and survival essentials, while the Medical Rescue module addressed the core aspects of the health emergency response. The Restore Work and Production module repaired the halt in production and livelihoods caused by the outbreak, strengthening and developing emergency coordination and roles across emergency organizations. The Implement Responsibility module provided more heterogeneous emergency response resources for the overall JPCM coordination, complementing the COVID cross-organizational emergency response coordination.Conclusion: The study on the JPCM case in China improves public health emergency management and aids informed decision-making.

Four-quadrant modeling may offer some constructive insights into the institutional diversity of the emergency responses to COVID-19. This study utilized a typological method to investigate institutional arrangements and the emergency management of epidemic responses in China. The task environment for emergencies was divided into four categories. Targeted policies were assigned to explicit task environments by placing them in one of the four quadrants: public health procedures, medical operation standards, supervisory and regulatory measures, and norms and instructions. Institutional diversity resulted when the very loci of decision-making carried a dominant or subordinate role, providing a hierarchical system for relating the institutional processes needed to address the challenges of institutional fragments.