Multigenerational health research using population-based linked databases: an international review

Area of Science:

• Public Health
• Epidemiology
• Biostatistics

Background:

• Family health history is a known risk factor for numerous health conditions.
• Systematic collection of multi-generational family health histories is complex.
• Routinely collected electronic databases offer potential for large-scale multigenerational health research.

Purpose of the Study:

• To compare population-based, multigenerational health research capabilities of linked databases across six international sites.
• To synthesize the attributes, benefits, and challenges of using these routinely collected data sources.

Main Methods:

• Systematic review and synthesis of published literature.
• Comparison of data linkage methods and data availability across European (Denmark, Norway, Sweden) and non-European (Manitoba, Taiwan, Western Australia) sites.
• Analysis of challenges in utilizing familial relationship and health condition data.

Main Results:

• European sites primarily use population registries for family structure identification; non-European sites use health insurance registries or linked multi-source data.
• Data availability for familial status varies, with Sweden having the earliest records (1947) and Taiwan the most recent (1995 onwards).
• All sites report near-complete coverage of familial relationships within their regions, but challenges remain in differentiating relationship types and ensuring data accuracy over time.

Conclusions:

• Routinely collected, linked databases are powerful tools for population-based multigenerational health research.
• Significant challenges exist in data linkage, accuracy, and differentiating biological versus legal familial relationships.
• Further research is needed to optimize the use of these data-intensive environments for advancing population health insights.