Improving bowel preparation for colonoscopy in the United Kingdom

Colorectal cancer (CRC) is diagnosed in 35 000 people in the UK each year, with 16 000 dying of the disease. Colonoscopy is the gold standard investigation of the bowel. Over 600 000 are performed each year in the UK. Removal of polyps is proven to reduce the subsequent incidence of CRC.
Bowel preparation is vital to good quality colonoscopy. Poor preparation is associated with missed lesions and an increased chance of subsequent cancer. Several scales have been devised to differentiate adequate from inadequate preparation. The Boston bowel preparation scale (BBPS) is the most validated scale. The quality of bowel preparation is dependent on numerous factors. Risk factors for poor preparation and modelling scores have previously been devised but have not been externally validated. Within this thesis is presented the results of a national survey of bowel preparation, giving an overview of practice in the United Kingdom (UK). “Can an Educational video improve the adequacy of BOwel Preparation for patients undergoing their first colonoscopy? (EBOPS)” examines whether a specific patient intervention with an educational video leads to an improvement in bowel preparation.
The national survey demonstrates that most units (93%) do not recommend patients to split bowel preparation for all colonoscopies, despite convincing evidence of the superiority of this timing regime. A correlation was seen with a significantly lower rate of inadequate preparation in those that advised split bowel preparation for colonoscopy.
The EBOPS study demonstrated that participants receiving access to an educational video were significantly more likely to have adequate bowel preparation (86.1% vs 79.1, p = 0.0471). Within this group, the oldest age cohort >70 year-olds derived the greatest benefit with a 13.9% absolute risk reduction, compared with 2.9% in <45-year-olds. Polyps were also significantly more likely to be found in the intervention group (39% vs 30%, OR 1.506, 95% CI 1.043 to 2.191, p < .032), demonstrating an improvement in resultant quality of the examination.
Interobserver reliability of BBPS grading conducted within a cohort of endoscopists from the EBOPS study demonstrated near perfect correlation. This supports the reliability of this grading scale and the findings of the EBOPS study, as well as highlighting the potential for the scale’s use within the UK.
Data from the EBOPS trial was used to validate two predictive models for poor bowel preparation. Inferior discriminant, calibration and descriptive values were demonstrated between this external validation data and the original developmental cohort. Although not as accurate as the original models at predicting poor bowel preparation, this validation analysis does indicate a minor degree of discrimination that can assist in stratification of risk of inadequate preparation.
This thesis identifies both a short fall in national practice, as well as a specific intervention that if adopted on a national scale, could lead to a significant improvement in bowel preparation at minimal cost.