gms | German Medical Science

Background: The centralisation of oesophago-gastric (O-G) cancer services in England was recommended in 2001, partly because of evidence for a volume-outcome effect for patients having surgery. This study (i) investigated the changes in surgical services for O-G cancer and postoperative mortality since centralisation. As most studies showing a volume outcome effect in resection surgery for oesophago-gastric cancer were conducted before the centralisation of clinical services we further (ii) estimated the volume-outcome relationship for both hospital and surgeon levels and (iii) assessed the robustness of this relationship for multiple outcome measures.

Research Question: Did the centralisation of oesophago-gastric (O-G) cancer services in England lead to the desired effects, as observed for different units of analysis (hospital vs surgeon) and different relevant outcome measures (mortality measured at different time intervals and anastomotic leakage rates)?

Methods: Patients with O-G cancer who had an oesophageal or gastric resection between April 2003 and March 2014 were identified in the national Hospital Episodes Statistics database. We derived information on the number of NHS trusts performing surgery, their surgical volume, and the number of consultants operating. Postoperative mortality was measured at 30 days, 90 days and 1 year. Logistic regression was used to examine how surgical outcomes were related to patient characteristics and organisational variables. In addition, using data from the National Oesophago-Gastric Cancer Audit from the UK, we fitted multivariable random-effects logistic regression models to quantify the effect of surgeon and hospital volume on three outcomes: 30-day and 90-day mortality and anastomotic leakage. The models included patient risk factors to adjust for differences in case-mix among hospitals and surgeons. The between-cluster heterogeneity was estimated with the median odds ratio (MOR).

Results: During the study period, 29 205 patients underwent an oesophagectomy or gastrectomy. The number of NHS trusts performing surgery decreased from 113 in 2003-04 to 43 in 2013-14, and the median annual surgical volume in NHS trusts rose from 21 to 55 patients. The annual 30 day, 90 day and 1 year mortality decreased from 7.4%, 11.3% and 29.7% in 2003-04 to 2.5%, 4.6% and 19.8% in 2013-14, respectively. There was no evidence that high-risk patients were not undergoing surgery. Changes in NHS trust volume explained only a proportion of the observed fall in mortality. Higher hospital volume was associated with lower 30-day mortality (OR: 0.94; 95% CI: 0.91-0.98) and lower anastomotic leakage rates (OR: 0.96; 95% CI: 0.93-0.98) but not 90-day mortality. Higher surgeon volume was only associated with lower anastomotic leakage rates (OR: 0.81; 95% CI: 0.72-0.92).

Discussion: Hospital volume explained a part of the between-hospital variation in 30-day mortality whereas surgeon volume explained part of the between-hospital variation in anastomotic leakage. In the setting of centralized O-G cancer surgery in England, we could still observe an effect of volume on short-term outcomes. However, the effect is inconsistent, depending on the type of outcome measure under consideration, and much smaller than in previous studies.

Practical Implications: Efforts to centralise O-G cancer services – and more widely, any efforts to centralise hospital services - should carefully address the effects of different units of analysis (hospital, team, surgeon volume) on a range of outcome measures that are relevant to patients.