Assessment and documentation of bowel preparation observed during colonoscopies is crucial in both research studies and real-world settings. In recent years, there have been an increasing number of studies that show that bowel preparation adequacy may serve as a covariate study comparing other outcomes, such as rates of adenoma detection or cecal intubation. Key factors associated with bowel preparation in colonoscopy-based research, which provides a thorough insight into the study designs, were highlighted in a recent review published in Gastrointestinal Endoscopy.1
When reviewing colonoscopy-based research studies, it is important to note the distinction between efficacy studies and clinical effectiveness studies. Efficacy studies have stricter inclusion and exclusion criteria, which may make the results less generalizable, especially when compared with clinical effectiveness studies, which are conducted under more real-world scenarios. For example, the Ottawa Bowel Preparation Quality Scale may be better suited for efficacy studies based on its components, whereas the Boston Bowel Preparation Score may be more appropriate for clinical effectiveness studies. The interval recall and whether or not it is altered on the basis of bowel prep is also another good indicator of clinical effectiveness in studies because it represents real-world scenarios. These types of factors, along with those mentioned here, should be closely reviewed with all physicians who will be performing the rating during the clinical trial. This may potentially help prevent any conflicts later in the study process.
In addition to efficacy and clinical effectiveness, there are other factors to consider, including the procedure times, time spent cleaning the colon during the colonoscopy, quantity of fluid aspirated, and volume of water used during the procedure. When adjunctive therapies such as water pumps, colonoscope oversleeves (Pure-Vu), and simethicone are used, evaluation of the bowel preparation can be performed both before and after the intervention. In addition, the ease of use and practicality of these interventions can be difficult to quantitate, and may vary in research settings, where physicians may have more time for colonoscopies compared with the real world. How effective an adjunctive therapy to bowel preparation can also be potentially detrimental to certain endpoints such as adenoma detection rate (ADR) based on physicians becoming “overconfident” and less careful with inspection when the bowel preparation is pristine2,3.
Dietary components of bowel preparation-based studies should also be reviewed based on how specific the dietary requirements are for patients (specific menu vs generalized information on low-fiber diet). The effect of a preprocedure diet also needs to be analyzed differently based on the timing (eg, morning vs afternoon) of the colonoscopy the following day.
“Real-time” evaluation of a bowel preparation by the endoscopist is frequently performed in studies. Before performing the colonoscopy, the endoscopist should receive standardized training so that there is less variability between physician ratings. Some studies will use prerecorded images analyzed by independent viewers at a later time to evaluate the bowel preparation. This can cause logistical issues with the institutional review boards and patient deidentification. In addition, many of these files can be relatively large and difficult to distribute. Still images are rarely used in place of video or live colonoscopies.
When reviewing bowel preparation scores during data analysis, some studies will “dichotomize” the scores into certain categories. For example, excellent and good vs fair or poor and Boston Bowel Preparation Score segment scores of 0 to 1 vs 2 and 3. This can have an effect on interrater reliability and variability. When independent raters of bowel preparation disagree, scores can be summarized using the mean, median, or total summation calculated, all of which could have significant effects on the final scores based on the type of scale used. A more costly (based on time and money) approach could be to have both reviewers review the colonoscopy together and come to a consensus.
Many bowel preparation studies opt for a noninferiority over superiority trial design, as these studies will typically require fewer patients and not aim to show a large difference over the comparator group. Superiority data can still be collected on certain secondary endpoints, and if a bowel preparation is superior within these, it could add a potential benefit in the eyes of a practicing gastroenterologist.
Conducting and evaluating research studies on bowel preparation for colonoscopies is challenging and is an area of significant interest both now and in the future. This review article helps outline some of these challenges and provides a unique way of evaluating published studies.
1. Jacobson BC, Calderwood AH. Measuring bowel preparation adequacy in colonoscopy-based research: a review of key considerations [published online September 27, 2019]. Gastrointestinal Endoscopy. doi:10.1016/j.gie.2019.09.031
2. Adike A, Buras MR, Gurudu SR, et al. Is the level of cleanliness using segmental Boston bowel preparation scale associated with a higher adenoma detection rate? Ann Gastroenterol. 2018;31(2):217-223.
3. Calderwood AH, Thompson K, Schroy III PC, Lieberman DA, Jacobson BC. Good is better than excellent: bowel preparation quality and adenoma detection rates. Gastrointest Endosc .2015;81(3):691-699.