Data Utility Framework
HDR UK has a mission to unite the UK’s health data to enable discoveries that improve people’s lives. This is being approached by bringing the data together through the work of the UK Health Data Research Alliance and Innovation Gateway (Uniting the Data), by making the data more useful for science and innovation through the work of the Health Data Research Hubs and the development of tools and approaches (Improving the Data) and using the data for specific purposes with the potential to transform people’s lives (Using the Data).
The work to Unite and Improve the Data is important, but it is only meaningful if it leads to greater quantity and quality of research and the generation of more meaningful insights. There is much discussion about the importance of data curation, and how much should be invested in this area. However, there is little clarity regarding exactly what activities are meant by ‘curation’ and in order to ensure that resources are effectively targeted, this should be informed by user needs. For example,
• A pharmaceutical company reviewing the effectiveness of a cancer treatment over a 10-year period requires linked primary and secondary data, with high levels of trust in the provenance and data quality management processes and follow-up of at least 10 years.
• A medical device regulator tracking adverse incidents associated with implantable devices requires detailed data on the devices and implantation procedure, linked device registry data with longitudinal patient information and consistency in coded records.
To achieve the ultimate aim of improving people’s lives, HDR UK has developed a framework to articulate the potential usefulness of datasets for specific purposes. This will support users of data in the discovery and selection of datasets for their purposes, as well as providing an evidence base for identifying specific areas of activity for improvement to allow for wider use or greater insights.
This has been developed in consultation with users of data and data custodians from a range of sectors, through interviews, surveys and a green paper consultation process1, working in partnership with MetadataWorks Ltd. The framework was further refined by being tested on approximately 50 datasets held by Health Data Research Hubs. The detail on how the framework was developed will be included in an upcoming academic publication.