Register-based research: some methodological considerations.

AIMS: Collecting new research data is often time consuming and expensive and it may be difficult to recruit volunteers to provide data for research. We should therefore make sure that we do not collect data that already are available and in the Nordic countries we have an opportunity to use national registers for certain types of research. In some situations using these population-based registers is preferable to collecting primary data since selection bias due to nonresponders is not a problem in these registers.
METHODS: Register-based data can be used in most designs available in observational studies and even add important data to randomised trials. Getting access to using the valuable data sources stored in national registers is much easier now that it was in the past. However, we still need better documentation and we should to a much larger extent be able to use research files that combine data from two or more of the Nordic countries. Linking primary research data to public registers will also provide new research opportunities.
RESULTS: Nordic registers have been used in several landmark papers over the years.
CONCLUSIONS: Register-based research can be done without risk of unwanted disclosure of personal data and it involves no invasive procedures. It is a valuable gift given by the people to be used for the people.