The national bowel cancer audit project: the impact of organisational structure on outcome in operative bowel cancer within the United Kingdom

Objectives: To investigate the relationship between organisational structure, process and surgical outcomes for bowel cancer surgery.

Methods: An e-survey was sent to the members of the Association of Coloproctology of Great Britain and Ireland to determine the organisational structure of their Trusts. Responses were combined with the National Bowel Cancer Audit (NBOCAP) data. Items investigated included; number of consultants, nurse specialists, volume of cases and intensive care facilities. Main outcome measures included: 30-day risk-adjusted mortality, length of stay (LOS), lymph node yield and circumferential margin involvement (CRM).

Results: One hundred and seventeen Trusts responded (65.8%), matched to 7666 patient episodes (NBOCAP data) from 54 (62.8%)Trusts who submitted data to the audit. Trusts treating <190 cases/annum (p > 0.001), <4 colorectal consultants (p > 0.001), <4 HDU beds (p > 0001) and <8 ITU beds (p > 0001) were more likely to have a 30-day-risk-adjusted mortality twice that of the national mean. Sixty five percent (n = 1603) of Trusts treating ≥ 190 cases/annum harvested ≥ 12 lymph nodes vs. 58.3% (n = 1435) in Trusts <190 cases/annum (p < 0.001). Trusts with ≥ 2 pathologists with an interest in bowel cancer harvested ≥ 12 lymph nodes more frequently (p=<0.001) and were more likely to identify extramural vascular invasion in the specimen (p = 0.015). Negative CRM was achieved in 81.4% (n = 81.4) of patients in Trusts treating ≥ 190 cases vs. 66.5% (n = 569) in Trusts<190 cases/annum (p < 0.001). Trusts offering fast track discharge were more likely to have a LOS < 15 days (p = 0.006). Surgeons treating ≤ 35 cases/annum had increased major post-operative complications (<35 cases = 70.2% vs. ≥ 35 cases = 21.9%; p < 0.001), however 30 day risk adjusted mortality was not increased in surgeons treating <35 cases/annum.

Conclusions: This study shows that the organisational infrastructure of hospitals appears to have as great an impact on patient outcomes as the volume of cases performed by hospital Trusts.