Variations in the height of tumour did not explain the variation in APE use. Specialist high-volume surgeons undertook fewer APEs and those they performed were closer to the dentate line than low-volume nonspecialist surgeons.

BACKGROUND: The National Health Service (NHS) bowel cancer screening programme (BCSP) was initiated across England in April 2006. To determine the feasibility of using national cancer registration data to assess the impact of the BCSP on stage-specific incidence, we studied trends in the incidence rates of colon (ICD10 C18) and rectosigmoid junction and rectum (ICD10 C19 -C20) cancers and the completeness of data on Dukes stage in England. METHODS: Data were obtained from all nine cancer registries for the period 1996 -2004, before the introduction of the BCSP, in men and women aged 50 -79 years. RESULTS: Overall, incidence rates declined by 1% per year in the 9 years before the introduction of the BCSP (Po0.001). Dukes stage was recorded for 60% of all registrations but this varied between regions and over time. Only four registries had completeness of 74% or more. Registrations with unknown Dukes stage decreased from 1996 to 2000, and then increased during 2001 -2004 affecting trends in stage-specific incidence. CONCLUSION: To study the impact of the BCSP on stage-specific incidence, regional variations in data completeness need to be addressed.

0.50, 0.80; 10 studies) (figure below), while 0.38 preferred TOE over SE (95% CI: -0.04, 0.80; 3 studies). Conclusion There is no difference between TOE and SE in terms of technical success rate and preference. Success rate of TNE <6 mm in diameter is equivalent to SE, but majority of patients prefer the former over the latter. Hence, TNE (<6 mm in diameter) should be the procedure of choice for screening. Introduction It is recognised that post-colonoscopy colorectal cancer (PCCRC) can be due to missed cancer, or cancer arising from missed or incompletely removed polyps. Thus the rate of post-colonoscopy colorectal cancer (PCCRC) should become a key quality indicator of colonoscopy. A quality indicator should be relevant to patients, clearly defined, standardised, and measurable over time and have a target to aim for. This study compares methods for defining PCCRC rates, proposes a method that best meets these criteria and explores rates over time.Methods Information on all individuals with a primary colorectal cancer and prior colonoscopic investigations in England between 2001 and 2010 was extracted from the National Cancer Data Repository. Previously published methods (Bressler, Cooper, Singh and leClerc) for deriving PCCRC rates were applied to these data to investigate the effect on the rate. A new method, based on the year of the colonoscopy, not CRC diagnosis, is proposed. Results Of 297,956 individuals diagnosed with colorectal cancer in the study period a total of 94,648 underwent a colonoscopy in the 3 years prior to their diagnosis. The table illustrates how application of the published methods and exclusion criteria to the dataset produces significantly different PCCRC rates from 2.4 to 7.8%: