Building a Longitudinal National Integrated Cardiovascular Database　- Lessons Learnt From SingCLOUD

Area of Science:

• Cardiovascular research
• Health informatics
• Real-world data analysis

Background:

• Fragmented real-world data hinders understanding of coronary artery disease (CAD) clinical outcomes and care quality.
• Existing data sources lack comprehensive longitudinal patient information.
• Need for integrated data to support robust cardiovascular research.

Purpose of the Study:

• To describe the rationale and design of the Singapore Cardiovascular Longitudinal Outcomes Database (SingCLOUD).
• To explain the approach for harmonizing diverse real-world data into a robust, longitudinal dataset.
• To present pilot data on myocardial infarction (MI) patients treated with percutaneous coronary intervention (PCI).

Main Methods:

• Developed a health data grid to integrate clinical, administrative, laboratory, procedural, prescription, and financial data.
• Harmonized data from diverse electronic medical and non-medical platforms across public healthcare institutions.
• Utilized pilot data from 3,819 patients with MI and PCI between 2012-2014.

Main Results:

• SingCLOUD platform generated 313 additional data fields compared to the Singapore Cardiac Data Bank.
• Identified 250 additional heart failure events and 664 major adverse cardiovascular events at 2 years.
• Captured 1-year low-density lipoprotein levels for 3,747 patients.

Conclusions:

• SingCLOUD enables in-depth analysis of real-world cardiovascular 'big data' by integrating multiple data sources.
• The platform enhances the depth and breadth of cardiovascular data analysis.
• Facilitates improved understanding of clinical outcomes and quality of care for CAD patients.