Introduction Colorectal cancer screening with colonoscopy has been shown to reduce mortality by removal of adenomatous polyps with potential for malignant change. Colonoscopists with higher adenoma detection rates have lower rates of interval cancer. The Bowel Cancer Screening Programme offers colonoscopy to subjects aged 60–74 after a positive faecal occult blood test (FOBT). All colonoscopists have to be screening-accredited with a lifetime log of over 1000 colonoscopies with certain minimum key performance indicators and then undergo a formal structured assessment process during which they are observed performing 2 colonoscopies by trained assessors.

Methods The aim was to assess how performance varied in this highly selected group of colonoscopists in the 12 months after they commenced screening. 139,363 procedures were performed between June 2006 and March 2012 by 245 colonoscopists. Data were collected on caecal intubation rate (CIR), the adenoma detection rate (ADR), the mean number of adenomas detected per patient (MAP) and the number of procedures performed by each colonoscopist.

Results Colonoscopists were divided into quartiles by performance for CIR and ADR. The mean CIR in the top quartile was 98.6% compared to 91.4% in the lowest quartile. The mean number of procedures in the first year of screening was 141 and 123 respectively (t-test p = 0.055); there was a correlation between the number of procedures performed and CIR (Pearson’s r = 0.196, p = 0.002). There was a significant difference in the number of procedures performed in the previous year before commencing screening (266 compared to 201, p = 0.02). There was no significant difference in the ADR (p = 0.48), pain caused, years of experience or number of lifetime procedures. There was no correlation between CIR and ADR (r = 0.001, p = 0.982). The mean ADR in the top quartile was 56.8% compared to 39.8% in the lowest quartile. There was no significant difference between these groups in the CIR, pain or any other measure from the lifetime procedure log. There was a significant difference in the MAP (1.64 compared to 0.77, p < 0.01). Overall there was a strong correlation between ADR and MAP (r = 0.59, p < 0.001).

Conclusion These data show that even in a stratified group of high-performing colonoscopists there is considerable variation in performance in CIR, ADR and MAP. There was no correlation between CIR and ADR but a significant association between CIR and the endoscopic activity the previous year. Colonoscopists performing more procedures in their first year screening patients did not have higher adenoma detection rates. Further studies are needed to ascertain the factors responsible for these differences to try to further improve performance and patient outcomes.

Disclosure of Interest None Declared.