National and regional health policies have increasingly emphasized the role of informatization in promoting integrity and compliance within public hospitals. The Large Hospital Inspection Work Plan (2023–2026) issued by China's National Health Commission mandates the establishment of risk prevention systems through digital means. In 2021, the Zhejiang Provincial Health Commission released the Evaluation and Management Measures for Clean Construction Index in Public Hospitals (Trial), providing a structured framework for assessing and monitoring ethical governance (6). More recently, the 2025 Work Priorities for Rectifying Unethical Practices in Medical Procurement and Services explicitly calls for “penetrative audit supervision” and end-to-end traceability of pharmaceuticals and medical consumables (7), reinforcing the need for integrated, intelligence-driven oversight.

Healthcare remains a high-risk sector for integrity violations due to complex operational environments, fragmented data systems, and traditional oversight limitations. Common challenges include data silos, delayed risk detection, and insufficient analytical capacity (8). Traditional manual supervision is inadequate for processing large-scale operational data in real time, creating an urgent need for intelligent systems that enable proactive risk management. The proposed platform addresses these gaps by offering dynamic monitoring, automated alerts, and data-driven decision support, thereby enhancing the precision and efficiency of compliance oversight (9).

This study employs a pre-post implementation comparison within a single institution. While the observed temporal associations are compelling, we acknowledge that factors such as concurrent national anti-corruption campaigns, institutional reforms, or leadership changes could have contributed to the outcomes. Therefore, the findings primarily demonstrate association rather than definitive causation. We further acknowledge three key limitations: First, the single-hospital design limits generalizability to diverse healthcare systems, institutional cultures, or regulatory contexts. Second, the absence of a control group prevents definitive attribution of outcomes solely to the platform, as concurrent regional health digital reform policies and national ethical governance initiatives may have synergistic effects. Third, the relatively short follow-up period (18 months) limits conclusions on long-term sustainability. Future multi-center studies with controlled designs and longitudinal follow-up (≥3 years) are recommended to strengthen causal inference and scalability. Additionally, the evaluation period (2024–2025) is relatively short for assessing long-term sustainability; plans for ongoing monitoring are in place.

The platform is structured around four core modules (Figure 1):

The platform integrates over 30 key performance and compliance indicators from multiple hospital subsystems. Using ETL (Extract, Transform, Load) pipelines for data integration and warehousing, heterogeneous data are cleansed, standardized, and stored in a structured data pool (10). To ensure security, compliance, and accountability, the platform implements a three-tier governance framework: (1) Data permissions: Role-based access control (RBAC) strictly limits sensitive data access (e.g., personal integrity files, individual prescribing data) to authorized staff in the Ethics and Compliance Office, with dynamic data masking for patient identifiers (11); (2) Privacy protection: A “data sandbox” is used for analytical queries, where raw data are desensitized (e.g., tokenization of patient IDs, aggregation of department-level metrics) in compliance with HIPAA (12) and China's Medical Data Security Management Measures (13); (3) Accountability mechanisms: The hospital's Ethics Committee conducts quarterly audits of platform operations, including data access logs and early-warning handling records, with annual performance reviews by a third-party professional institution (14). The Hangzhou Municipal Health Commission retrieves and aggregates relevant data through the platform interface, and accesses early-warning information and operational reports to ensure regulatory alignment.

Following the platform's launch in July 2024, integrity risk incidents decreased from 5 in the first half of 2024 (baseline) to 2 in the second half of 2024, representing a reduction of 60% (Rate Ratio, IRR = 0.40; 95% CI: 0.08–2.09). The downward trend continued in the first half of 2025, with only one incident recorded—a further 50% reduction compared to the initial post-implementation period (IRR = 0.50; 95% CI: 0.05–5.57). Most notably, during the second half of 2025, no integrity risk incidents were recorded, achieving a 100% reduction compared to the baseline and resulting in an rate of zero (95% CI: 0.0–3.7 per half-year). Although the confidence intervals are wide due to the low absolute number of events—a reflection of the hospital's overall low baseline risk—the consistent, stepwise reduction culminating in zero incidents over 18 months provides strong evidence of the platform's sustained, real-world effectiveness in risk prevention (Table 1).

Medical disputes fell from 4 cases in 2024 to 1 case in 2025, a 75% year-on-year reduction (Table 2), reflecting improved procedural standardization and communication transparency.

Patient satisfaction indicators showed consistent improvement post-implementation (Table 3). The platform complaint resolution rate increased from 95.1% to 97.3%, while discharge follow-up issue rates declined by 19.7%.

A critical consideration in implementing an integrity supervision platform is its potential impact on staff psychological safety and the organizational culture (21). To avoid creating an atmosphere of punitive surveillance, our platform was designed with principles of supportive governance and a just culture in mind. Thus, the platform redefines traditional supervision as a form of enabling governance and systemic support, aiming to prevent errors rather than merely punish them.

The case of Hangzhou Xixi Hospital demonstrates that such platforms are not merely technological upgrades but strategic tools for systemic governance reform. Future developments may incorporate machine learning for predictive analytics, potentially linking integrity monitoring with direct clinical safety indicators (e.g., medication error patterns, diagnostic delay alerts) to further bridge ethical governance and patient outcomes, blockchain for tamper-proof audit trails, and interoperable designs for multi-institutional data sharing.