Scottish Administrative Data Research Partnership

The aim of the partnership is to enable research that leads to policy decisions that will in turn will help Scotland progress towards the vision outlined in the National Performance Framework. This framework helps to shape high level research priorities for Scottish Government, including tackling poverty, providing quality jobs and fair work for all, and ensuring that we live in inclusive, empowered, resilient and safe communities. S-ADRP consists of a number of Strategic Impact Programmes (SIPs) each dealing with a research priority.

Underpinning all this is data. EPCC’s role in the partnership will be to help provide research-ready data from the numerous sources that will be needed to support policy-making decisions. These sources include health  and educational records, police and judicial databases, census data and emergency services data. Of course, there are governance issues related to the use of this data, as it comes from many sources and contains sensitive information. The framework that will allow researchers to use this data must ensure that legal and ethical practices (such as the removal of data that enables identification of an individual) are followed.

The technical infrastructure that will support this is a Safe Haven operated by EPCC, which protects patient identity and privacy while allowing data from electronic records to be used to support research when it is not practicable to obtain individual patient consent.

The partners are currently investigating ways to link the data derived from various databases so that it can be used by researchers working in the SIPs. This linked data must provide the information needed by the researchers to derive useful findings, but must also preserve the privacy of individuals according to the governance policies.

In addition to this is the task of making the data ready for research. This includes ‘cleaning’ the data (for example identifying and dealing with records containing data that is clearly out of range) and, since the databases will be large, optimising them to support the types of query that the researchers will generate.

You can read more about our data work in our Research webpages.