Understanding the Integrated Health Management System Policy in China From Multiple Perspectives: Systematic Review and Content Analysis.

BACKGROUND: The integrated health management system (IHMS), which unites all health care-related institutions under a health-centered organizational framework, is of great significance to China in promoting the hierarchical treatment system and improving the new health care reform. China's IHMS policy consists of multiple policies at different levels and at different times; however, there is a lack of comprehensive interpretation and analysis of these policies, which is not conducive to the further development of the IHMS in China. OBJECTIVE: This study aims to comprehensively analyze and understand the characteristics, development, and evolution of China's IHMS policy to inform the design and improvement of the system. METHODS: We followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines to collect 152 policy documents. With the perspective of policy tools and policy orientation as the core, a comprehensive 6D framework including policy level, policy nature, release time, policy tools, stakeholders, and policy orientation was established by combining the content of policy texts. These dimensions were then analyzed using content analysis. RESULTS: First, we found that, regarding the coordination of policy tools and stakeholders, China's IHMS policy was more inclined to use environment-based policy tools (1089/1929, 56.45%), which suggests a need for further balance in the internal structure of policy tools. Attention to different actors varied, and the participation of physicians and residents needs further improvement (65/2019, 3.22% and 11/2019, 0.54%, respectively). Second, in terms of level differences, Shanghai's IHMS policy used fewer demand-based policy tools (43/483, 8.9%), whereas the national IHMS policy and those of other provinces and cities used fewer supply-based tools (61/357, 17.1% and 248/357, 69.5%, respectively). The national IHMS strategy placed more emphasis on the construction of smart health care (including digital health; 10/275, 3.6%), whereas Shanghai was a leader in the development of healthy community and healthy China (9/158, 5.7% and 4/158, 2.5%, respectively). Third, in terms of time evolution, the various policy tools showed an increasing and then decreasing trend from 2014 to 2021, with relatively more use of environment-based policy tools and less use of demand-based policy tools in the last 3 years. The growth of China's IHMS policy can be divided into 3 stages: the disease-centered period (2014-2017), the e-health technology development period (2017-2019), and the health-centered period (2018-2021). CONCLUSIONS: Policy makers should make several adjustments, such as coordinating policy tools and the uneven relationships among stakeholders; grasping key policy priorities in the context of local characteristics; and focusing on horizontal, multidimensional integration of health resources starting from the community. This study expands the objects of policy research and improves the framework for policy analysis. The findings provide some possible lessons for future policy formulation and optimization.

Artificial Intelligence-Based Multimodal Risk Assessment Model for Surgical Site Infection (AMRAMS): Development and Validation Study.

BACKGROUND: Surgical site infection (SSI) is one of the most common types of health care-associated infections. It increases mortality, prolongs hospital length of stay, and raises health care costs. Many institutions developed risk assessment models for SSI to help surgeons preoperatively identify high-risk patients and guide clinical intervention. However, most of these models had low accuracies. OBJECTIVE: We aimed to provide a solution in the form of an Artificial intelligence-based Multimodal Risk Assessment Model for Surgical site infection (AMRAMS) for inpatients undergoing operations, using routinely collected clinical data. We internally and externally validated the discriminations of the models, which combined various machine learning and natural language processing techniques, and compared them with the National Nosocomial Infections Surveillance (NNIS) risk index. METHODS: We retrieved inpatient records between January 1, 2014, and June 30, 2019, from the electronic medical record (EMR) system of Rui Jin Hospital, Luwan Branch, Shanghai, China. We used data from before July 1, 2018, as the development set for internal validation and the remaining data as the test set for external validation. We included patient demographics, preoperative lab results, and free-text preoperative notes as our features. We used word-embedding techniques to encode text information, and we trained the LASSO (least absolute shrinkage and selection operator) model, random forest model, gradient boosting decision tree (GBDT) model, convolutional neural network (CNN) model, and self-attention network model using the combined data. Surgeons manually scored the NNIS risk index values. RESULTS: For internal bootstrapping validation, CNN yielded the highest mean area under the receiver operating characteristic curve (AUROC) of 0.889 (95% CI 0.886-0.892), and the paired-sample t test revealed statistically significant advantages as compared with other models (P<.001). The self-attention network yielded the second-highest mean AUROC of 0.882 (95% CI 0.878-0.886), but the AUROC was only numerically higher than the AUROC of the third-best model, GBDT with text embeddings (mean AUROC 0.881, 95% CI 0.878-0.884, P=.47). The AUROCs of LASSO, random forest, and GBDT models using text embeddings were statistically higher than the AUROCs of models not using text embeddings (P<.001). For external validation, the self-attention network yielded the highest AUROC of 0.879. CNN was the second-best model (AUROC 0.878), and GBDT with text embeddings was the third-best model (AUROC 0.872). The NNIS risk index scored by surgeons had an AUROC of 0.651. CONCLUSIONS: Our AMRAMS based on EMR data and deep learning methods-CNN and self-attention network-had significant advantages in terms of accuracy compared with other conventional machine learning methods and the NNIS risk index. Moreover, the semantic embeddings of preoperative notes improved the model performance further. Our models could replace the NNIS risk index to provide personalized guidance for the preoperative intervention of SSIs. Through this case, we offered an easy-to-implement solution for building multimodal RAMs for other similar scenarios.