International Multi-database study PREparedness

Leveraging Local Healthcare Data for Global Real-World Evidence

Developed by the Population Health Data Center (PHDc) at National Cheng Kung University, the IMPRESIVE framework harmonizes local health data into timely, multi-country evidence for faster healthcare decision-making.

See how IMPRESIVE works→ View case studies
National Cheng Kung University
Population Health Data Center

What is IMPRESIVE?

IMPRESIVE stands for International Multi-database study PREparedness: Databases Standardization, Integration and Visualization for timely Evaluation on Disease and Treatment.

IMPRESIVE serves as a research infrastructure designed to facilitate international multi-database studies across four critical domains: data, methods, collaboration, and interpretation.

The IMPRESIVE study workflow

Walk through the architecture

  1. Research question

    Translating clinical, policy, or regulatory questions into robust data analysis approaches. Why preparedness starts here →

  2. Feasibility & fit for purpose

    Evaluating whether local data environments can support the intended analytical strategy. Criteria →

  3. Protocol & variable definition

    Harmonizing definitions across data centers, including analysis plans, code lists, variable specifications, and output formats. Code lists →

  4. Standardization

    Applying Common Data Model structures, such as OMOP or Sentinel-like frameworks, to establish a unified analytical interface tailored to local settings. Routes and mappings →

  5. Quality validation

    Verifying structural conformance, completeness, plausibility, and study-specific validity of converted datasets. Quality framework →

  6. Distributed execution

    Executing standardized analytical programs locally within each data center while documenting site-specific adaptations, exceptions, and limitations. What travels →

  7. Evidence integration

    Aggregating site-level outputs while preserving site-level context. Case portfolio →

  8. Visualization & interpretation

    Presenting country-specific and integrated findings through dynamic visual analytics, clearly highlighting uncertainty and contextual insights. Visualization platform →

Key figures

Data environments
6
National claims and hospital electronic records across Taiwan, South Korea, Japan, and Hong Kong.
Standardization routes
2
OMOP and Sentinel common data models.
Cohort sample size
⁓3M
ASCVD, Atopic dermatitis and Prurigo nodularis cohorts.

Accomplishments of the IMPRESIVE Framework

Explore how our framework has driven international health research and supported multi-country real-world evidence studies.

Case 01 · ASCVD

Risk patterns across East Asian data environments

11.21.41.6NHIRD1.51NHIS1.47CGRD1.49DeSC1.17CDARS1.16HR (95% CI), ASCVD recurrence
EstablishCross-database comparability
Read more →
Extended case · ASCVD visualization

Explore standardized aggregate outputs interactively

051012.47.8ASCVD recurrence9.75.8Major CV events3.31.6ACSVery-high-riskHigh-riskIncidence per 100 person-years
ExploreInteractive output review
Read more →
Case 02 · AD / PN

A reusable framework across therapeutic areas

0%25%50%D1D2D3NHIRDHIRADeSCCGRDAdult AD prevalence, 2020
ExpandReusable study machinery
Read more →
Case 03 · Transportability

Can evidence travel from Taiwan to Japan?

0.60.811.21.4TW obs.0.81Transported0.95JP obs.1.12RR (95% CI), MACE · prasugrel vs clopidogrel
ExtendEvidence transportability
Read more →

Need a Data Partner for Your Next Case Study?

Share your research goals and data requirements with us. Let’s collaborate to advance your real-world evidence investigation.

Join the conversation