Three Key Considerations for Hospitals Procuring CDSS: Insights from U.S. Experts

Clinical Decision Support Systems (CDSS) aim to ultimately improve the quality and efficiency of healthcare by influencing physicians’ behaviors and decisions. As clinical big data continues to accumulate, CDSS is becoming a core information system in hospitals and an important tool for quality management and enhancing comprehensive institutional capabilities. Since the National Health Commission issued the “Notice on Further Promoting the Construction of Information Systems Centered on Electronic Medical Records in Healthcare Institutions,” which has placed further requirements on electronic medical record informatization and clinical decision support functionalities, the application of CDSS will continue to expand.

To achieve the “2020” goals for electronic medical record (EMR) informatization, major hospitals are accelerating the overhaul of their information systems and strengthening the deployment of clinical decision support systems (CDSS). Faced with CDSS products that vary in technology, sophistication levels, and application scenarios, how can hospitals select the “most suitable” one? Experts in the field of healthcare informatics in the United States have offered CIOs some tips and recommendations to address these specific challenges.

Kody Hansen, Research Director at KLAS, a well-known medical IT data research firm, stated that healthcare organizations’ CMOs and CIOs have occasionally reported wasted human resources due to the lack of a clearly defined implementation timeline upfront. “A simple question during the procurement process could save you hundreds of hours,” he said. Implementing a Clinical Decision Support System (CDSS) requires robust teamwork between vendors and healthcare organizations, as CDSS typically needs to be integrated with other information systems, such as Electronic Health Record (EHR) systems.

Therefore, Kody Hansen recommends that CIOs review vendors’ prior case studies to evaluate measurable implementation timelines, the standardization of knowledge bases, system scalability, and content customizability for upcoming CDSS deployments. Adopting products with high scalability and more comprehensive knowledge bases can help reduce implementation burdens. Huimei CDSS employs a distributed framework that seamlessly integrates into existing hospital information system platforms without requiring system modifications. It enables rapid interface response times and delivers high-concurrency, high-availability services, and has been clinically deployed in nearly 60 large general hospitals.

Patients’ medical record data are central to clinical decision support technology and serve as the “atoms” of hospital information systems; thus, data structuring and standardization are critical. The adoption of unified, standardized data structures by CDSS vendors is a prerequisite for healthcare institutions to unlock the “boundless potential” of their data and enable data sharing and portability.

Another key focus is the seamless integration of CDSS into healthcare professionals’ workflows to support clinical operations. The AI-powered Huimei CDSS leverages the hospital information platform for data collection, integrating data from various systems including HIS, LIS, EMR, and CDR. It employs real-time stream computing to clean and perform terminology mapping on multi-source heterogeneous data, thereby constructing a standardized database. On this foundation, it supports clinical big data mining, machine learning, and decision support, while enabling interactive features and passive alerts during physicians’ electronic medical record documentation, thus meeting the requirements for coverage across the entire spectrum of hospital operations.

As the value of big data increases, clinical decision support systems (CDSS) have gone beyond simple “alert and reminder” functions, delivering substantial economic and social value. However, for CIOs, implementing these systems requires careful consideration of the question: “Where lies the greatest value of CDSS for our hospital?”

Kody Hansen stated that CDSS is neither “omnipotent” nor “useless”; whether system deployment proves to be a “highlight” or a “burden” often depends on the organization’s philosophy. For CIOs, properly evaluating CDSS requires consideration of the following factors:

1) Is the adoption of CDSS by hospitals a short-term necessity or driven by long-term potential? Meeting regulatory or third-party certification requirements is currently the most direct motivation for hospitals, representing a lower-level need. While many products can temporarily fulfill these requirements, hospitals should exercise caution with “one-off” purchases.

2) If the goal of introducing clinical decision support is to “reduce negative alerts, provide reminders within end-users’ workflows to better monitor adverse events, and promote improvements in evidence-based diagnostic and treatment decisions,” greater resources and patience must be invested.

3) You must be convinced that clinical decision support systems are a “value-add” component in building a value-based healthcare system.

“Always keep your core objectives and users in mind, and ensure that you meet their needs,” emphasized W. Edward Reynolds, Chief Technology Officer and Vice President of a leading healthcare IT solutions provider. He highlighted that the key to successful CDSS deployment is not always a technical issue; rather, the content (alerts) and its delivery are paramount. A truly “useful” advanced CDSS must possess three capabilities simultaneously: real-time data management, algorithmic models, and integration with clinical workflows—particularly regarding the timeliness of information processing and the ability to provide timely alerts at appropriate decision points.

In China, Huimei CDSS has developed and designed an intelligent clinical decision support system centered on improving clinical quality, leveraging natural language processing and deep learning technologies. By constructing deep learning models based on the Mayo Clinic knowledge framework and electronic medical record data, the system assists healthcare professionals in differential diagnosis, optimizes treatment plans, and manages clinical quality, thereby effectively enhancing the consistency and standardization of clinical practices. It has been widely deployed in nearly 60 large hospitals and hundreds of primary healthcare institutions, providing comprehensive support for clinical quality management and the advancement of intelligent electronic medical records.