Introduction Clinical diagnosis and procedure coding for hospital admissions is mandatory in order to receive payment through the NHS tariff system. This does not extend into outpatient attendances where, to date, there has not been a nationally agreed process for coding. In order to address this, Connecting for Health (superseded by the Health and Social Care Information Centre), in 2013, invited clinical subspecialty societies to submit codes (limited to approximately 100) that would encapsulate 90% of their outpatient diagnoses. In this study we sought to validate the BSG hepatology subset* to see if it a) met the scope criteria for HSCIC and b) could be clinically useful in outpatient resource planning.

Methods We have studied the outpatient clinic letters in patients seen by hepatologists or hepatology nurse specialists in the Gloucestershire outpatient departments. A spreadsheet with datafields for postcode, diagnoses and SNOMED mappings within the BSG dataset was developed. Microsoft MapPoint® was used to examine the geographic distribution of the cohort with the intention of improving follow-up arrangements for patients in relation to the location of community hospitals.

Results From 236 patients studied to date, 100% of patients could be assigned a code from within the dataset. However, 32.2% had alternative/co-existing diagnoses that could also be legitimately coded. In order to be clinically useful the outpatient record should contain datafields for both aetiology (specific) and stage of liver disease (generic). For example, the MapPoint exercise provided an insightful distinction between requirements for a viral hepatitis or cirrhosis clinic in community hospitals.

Conclusion The BSG hepatology dataset satisfies the scoping requirements of HSCIC but a single diagnostic datafield entry is not immediately useful to clinicians, service providers or commissioners since treatment pathways in terms of aetiology and management pathways in terms of stage need not correlate.

Disclosure of Interest None Declared.