Quality outcomes comparison of total thyroidectomy between UKRETS and HES databases

Abstract

Abstract Aims Hospital Episodes Statistics for England (HES) and UK Registry of Endocrine and Thyroid Surgery (UKRETS) are the largest databases on thyroidectomy within the UK. A validation of UKRETS data accuracy has never occurred. The purpose of this study was to compare quality outcomes in total thyroidectomy (TT) between the UKRETS and HES datasets. Methods A retrospective comparison study between UKRETS and HES for TT from April 2012–March 2018. Vocal cord palsy, hypocalcaemia, mortality & length of stay were the quality outcomes studied. Results The HES and UKRETS databases contained 23,025 and 11,242 patients respectively with UKRETS capturing 48.8% of HES data for TT. 18.8%(42/223) of surgeons performed ≥20 procedures/year in 54.3%(6110/11,242) of UKRETS procedures compared to only 9.7%(72/740) of surgeons from HES performing 44.6%(10,279/23,035) of all TTs. Median length of stay was significantly higher within in HES(2 days, IQR 1-3) versus UKRETS(1 day, IQR 1-2). 30.7%(7,071) of HES patients stayed for >2 days compared to 18.7%(1,837) within the UKRETS registry. Both databases reported a mortality of 0.1%. Post-operative hypocalcaemia was reported higher in UKRETS (21.7%) compared to HES (6.8%), p<0.0001. Rates of voice change were significantly higher in UKRETS (9.4%) compared to those in HES (1.3%), p<0.0001. Conclusions Surgeons contributing to UKRETS had a higher volume practice and shorter stay than those within HES. UKRETS appears to underreport the number of procedures conducted, with less low volume surgery captured. However, recording of hypocalcaemia and voice changes is mandatory in UKRETS, and HES appear to substantially under-report such complications.