Introduction Colonoscopy frequently causes discomfort and a range of medications are used to improve the patient experience. The relationship between medication use and patient comfort, however, is complex and subject to a number of potential biases. We sought to describe the relationship between patient comfort and medication use within the English Bowel Cancer Screening Programme (BCSP).

Methods Procedural information for colonoscopy examinations performed within the English BCSP is prospectively entered into a national database. Comfort is independently rated by a specialist screening practitioner (SSP) using the Modified Gloucester Comfort Scale (no, minimal, mild, moderate and severe). We studied significant patient discomfort (moderate or severe) and medication usage for colonoscopists performing over 100 examinations between January 2010 and December 2012. Comparisons were made using the χ2 test and correlations were analysed using Spearman rank correlation coefficient.

Results During the period of the study 113,316 examinations were performed by 290 endoscopists. Significant discomfort occurred during 8.9% of colonoscopy examinations but there was variation between individual colonoscopists (median 8.1%, IQR 5.0–12.6%, range 0.8–23.9%). Significant discomfort was more common in females (12.7 vs. 6.1%, odds ratio (OR) 2.24), patients with diverticulosis (11.8% vs. 8.7%, OR 1.34), incomplete examinations (37.3 vs. 7.9%, OR 6.8), inadequate bowel preparation (13.5 vs. 9.6%, OR 1.4) and screening rather than surveillance colonoscopies (9.1 vs. 7.4%, OR 1.24). Midazolam was administered during 87.8% and opiate analgesia during 87.3% of procedures. There was wide variation between colonoscopists in the proportion of examinations in which midazolam (median use 95.1%, IQR 81.8–97.8%, range 4.1% >100%) and opiate analgesia (median use 97.3, IQR 85.0–99.2%, range 5.6–100%) were used. Reversal agents were rarely used (8 in 10,000). Entonox was administered during 7.5% of examinations but most who administered it did so in a minority of their procedures (median use 0.7%, IQR 0–8.2%, range 0–98.9%). 4.7% of patients underwent medication-free colonoscopy. General anaesthesia was rarely used (0.5%). There were no significant correlations between the amount or proportion of medication used by colonoscopists and the comfort of their patients.

Conclusion Most colonoscopy examinations were performed without causing significant discomfort. Although most colonoscopists used intravenous medication those who used less medication were no more likely to cause significant discomfort. Appropriate use of medication to achieve comfortable procedures while minimising risk and inconvenience remains an important focus for future research.

Disclosure of Interest None Declared.