Data quality during processing

The HES Data Quality team investigate the quality of data and produce a number of reports, throughout the processing of the HES dataset. This enables the team to identify and remedy any specific issues and to check that the automatic cleans that run during processing are performing correctly.

Checking and reporting of data quality during processing includes:

• Investigating the national monthly activity in the initial SUS extract used to create HES, in comparison to previous extracts to determine completeness of the data
• Reporting the use of obsolete or invalid provider codes, to be mapped to that provider's new successor codes
• Examining the level of population of all fields, highlighting changes in population that could indicate field specific quality issues
• Detailed reporting of records flagged as potential duplicates to identify whether they should be removed from HES
• Identifying providers with missing data that was present in previous extracts, who may have not submitted or overwritten recent data. Issues can be noted and providers contacted.

Further information about the cleans that are performed on HES data based on these checks is available in the Data Cleaning section.