Background Suboptimal bowel preparation has several consequences, including reduced polyp detection rate and increased cost of colorectal cancer screening. The presence of constipation is thought to be a feature associated with poor bowel preparation.

Objectives To characterise the relationship between features of constipation and the quality of bowel preparation during colonoscopy.

Design Patients presenting for an outpatient colonoscopy were asked to complete a questionnaire which included demographics, type of bowel purgative and features of bowel movements (BMs)—derived from the ROME III criteria for diagnosis of chronic constipation.

Patients 101 patients from the community undergoing surveillance colonoscopy completed the study.

Interventions Patients underwent standard bowel preparation, completed a pre-endoscopy survey, followed by routine surveillance colonoscopy.

Main outcome measurements The endoscopist rated the quality of bowel preparation using the previously validated Ottawa scoring scale. Statistical analysis was performed to characterise the relationship between the existence of chronic constipation and quality of bowel preparation.

Results BM frequency of <1/day, 1/day, 2/day, or 3/day was inversely correlated with average total Ottawa score (range 5.93 to 4.00), p=0.028) as well as recto-sigmoid Ottawa score (range 1.8 to 1.0, p=0.006). Among women, there was a statistically significant (p=0.025) association between those who reported hard stools (Ottawa=6.3) and those who denied hard stools (4.5).

Conclusion Eliciting features of BMs before colonoscopy may predict those at risk for poor bowel preparations and increase cost effectiveness of colonoscopy.