Not a collection of disconnected outputs — an infrastructure built iteratively, study by study, since the earliest multinational collaborations.
Demonstrated in practice
Real studies, run on the shared infrastructure, across multiple countries and two common data models.
Compared subsequent events between patients at very high versus high risk of recurrence — chosen partly because the clinically expected direction provided a built-in benchmark for the data transformation and analysis, and partly to test whether the 2018 AHA/ACC risk-stratification framework, developed in Western populations, retains discriminatory value in Asia.
Examined prevalence, standard of care, medication use and clinical outcomes — comparing case definitions of increasing stringency, from any single diagnosis to repeated diagnoses within 180 days combined with systemic medication criteria (with a Hong Kong-specific adaptation).
The toolkit that accumulated
Prevalence, incidence, baseline covariates, outcome estimation and subgroup analysis — parameterized so new results can be generated rapidly.
OMOP and an adapted Sentinel-like model, each with transformation programs, mapping tables and quality checks.
Cross-country visualization of results, built on the standardized outputs of the shared programs.
Transportability analyses, Taiwan–Japan methodological comparisons, and a documented drug-dose table extending the Sentinel design.
Because transformation processes and shared programs already exist, new multi-database study ideas can be assessed, adapted and expanded to additional sites without rebuilding the workflow.
Our story
Taiwan, Hong Kong and Korea used a common protocol to study the utilization patterns and effectiveness of bone-targeting agents in breast, prostate and lung cancer with bone metastasis. Even with a shared protocol, substantial differences arose from variation in healthcare systems, data content and programming practices — the lesson that sparked IMPRESIVE.
Introduced the OMOP common data model to study baseline characteristics and recurrence risk in hospitalized ASCVD patients — with vocabulary mapping and quality assurance developed on structure-identical dummy data for restricted data centers.
Adapted the Sentinel CDM to test result consistency across CDMs and follow-up periods, then extended into atopic dermatitis and prurigo nodularis — prevalence, baseline characteristics and prescribing patterns — with a growing focus on visualization.
Collaboration across Taiwan's NHIRD and CGRD, Korea's NHIS and HIRA, Hong Kong's CDARS, Japan's DeSC and the UK's CPRD. Around 15 CDM disease cohorts now stand across countries and disease areas — from tens of thousands to millions of patients each.
Multinational visualization dashboards, transportability analyses and advanced methodological comparisons — built on the established transformation processes and shared programs.
New countries, databases, methodological modules and research outputs are incorporated over time. The unfinished work is the invitation.
Why it matters
Distinguishing what has been achieved, what is emerging, and the value still to be created.
Improved reproducibility in multinational research, capacity to study rare events and subgroups, comparisons of CDM suitability, identification of cross-country consistency and heterogeneity, and reusable methods and tools.
Real-world evidence from multiple countries supporting clinical guidelines, regulatory decisions, drug safety and health policy in a timely manner — without extrapolating a single Western population to all of Asia.
Common data models, modular programs and distributed execution shorten study initiation and data-processing time, cut the cost of rebuilding workflows, and support drug development and post-marketing surveillance.
Shared standards, tools and training help researchers in Thailand, South and Southeast Asia and beyond build the capacity to generate their own real-world evidence — impact beyond publication counts.
What we produce
Peer-reviewed multi-country studies and methods papers.
Presentations at scientific symposia and meetings.
Decision-oriented reports for regulators, HTA and policy.
Reusable designs, phenotypes and analysis modules.
Courses, slides and workshop content for capacity building.
Readiness, results and evidence-to-decision dashboards.
Pre-specified protocols with open, version-controlled code.
Concise evidence-to-decision summaries for stakeholders.
Current pipeline: the East Asia ASCVD manuscript is under submission, the CDM-concordance manuscript is in drafting, and the VESALIUS-REAL abstract has been accepted at APSC 2026. A consolidated list of published outputs will be added here as items go public, verified against project records.
Propose a cross-national question and build on everything that is already here.
Submit a proposal