Endoscopy 2018; 50(01): 4-5
DOI: 10.1055/s-0043-121146
© Georg Thieme Verlag KG Stuttgart · New York

Colonoscopy quality: measuring the patient experience

Referring to Valori RM et al. p. 40–51Jennifer Nayor1, 2, John R. Saltzman1, 2
  • 1Division of Gastroenterology, Hepatology and Endoscopy, Brigham and Women’s Hospital, Boston, Massachusetts, United States
  • 2Harvard Medical School, Boston, Massachusetts, United States
Further Information

Publication History

Publication Date:
21 December 2017 (online)

Performing high quality, safe colonoscopies is essential for good patient care and prevention of missed or interval colon cancers [1]. Primary quality metrics include the adenoma detection rate (ADR), cecal intubation rate, and bowel preparation adequacy [2] [3]. The United Kingdom’s multisociety working group guidelines also recommend tracking median doses of sedation and patient comfort levels [2].

“...the use of the PICI composite score for quality and safety of cecal intubation has some pitfalls, including the potential lack of applicability to other countries with different patient populations and sedation practices.”

The study presented by Valori et al. in this issue of Endoscopy introduces a new colonoscopy performance indicator, the Performance Indicator of Colonic Intubation (PICI), which is a composite measure of cecal intubation rate, patient comfort, and dose of sedation [4]. PICI was developed using data from the 2011 UK national colonoscopy audit [5]. PICI was achieved in half of the colonoscopies. PICI was associated with more subgroup variables than cecal intubation alone and cecal intubation with acceptable comfort level, which the authors conclude makes this metric better able to detect differences in performance. Procedures in which PICI was achieved were more likely to identify polyps.

The new quality indicator introduced in this paper has the benefit and novelty of including patient-centered metrics. The primary colonoscopy quality indicators, ADR, cecal intubation, and withdrawal time, are endoscopist-dependent variables that correlate with interval or missed colon cancer rates, but do not evaluate the patient’s experience during colonoscopy. Patient comfort during colonoscopy should lead to safer procedures with less risk of perforation, and make patients more amenable to undergoing repeat procedures.

Patient comfort is challenging to measure. The Valori et al. study uses the Gloucester comfort scale, which is a 5-point scale based on nurse-reported intraprocedure comfort level. In the initial validation of this score, nurse-related comfort level was strongly correlated with the patient’s postprocedure evaluation of their experience (r = 0.92, P > 0.001) [6]. However, three-quarters of patients with a poor nurse-reported comfort score (4 – 5 out of 5) rated their colonoscopy experience as better or as expected, which suggests that the Gloucester comfort score overestimates some patients’ perception of discomfort.

The authors chose midazolam 2 mg as the cutoff in the PICI score. This dose was chosen as it was the median dose of midazolam used in the UK national colonoscopy audit. Many patients will need more sedation and going above this threshold should not imply fault in the endoscopist. A study looking at sedation trends in over 1 million colonoscopies performed in the United States, showed that younger, white, female patients, and colonoscopies with interventions performed required higher doses of midazolam [7]. In that study, the average dose of midazolam was 4 mg. Sedation doses and patients’ expectations of sedation may vary by country, as evidenced in the differing mean sedation doses in these studies. Patient populations may vary by endoscopist and by the area in which the colonoscopy is performed, which will influence PICI.

Propofol sedation for colonoscopy is widely utilized in many countries and is responsible for about half of sedation cases currently in the United States. The 2011 UK colonoscopy audit showed that the use of propofol varied by county from 0.3 % in England and Scotland up to 2.4 % in Northern Ireland [5]. Patients may prefer the deeper sedation afforded using propofol as well as the rapid recovery from sedation. Regional sedation practices will make implementation of the PICI score difficult and limit its generalizability.

Using a composite score easily summarizes data points that are highly related to each other, but can also overshadow the individual components of the score. The initial validation of the Gloucester comfort score reported a strong correlation between nurse-related comfort level and cecal intubation rate and average sedation dose [6]. In contrast, data from the UK screening program showed wide variation in the use of sedation with no association between the amount of sedation used and significant discomfort [8]. This finding highlights the subjective nature of a nurse-reported comfort score. Using a binary outcome for amount of sedation is limiting. A point system for amount of sedation used may give a better representation of quality and safety given the wide range of sedation doses used in practice. However, the use of a quality indicator that rewards the physician for giving less sedation may lead to undersedation of some patients by physicians trying to meet the quality metric.

The gold standard of colonoscopy quality is interval colon cancer, and a strongly correlated surrogate colonoscopy quality measure is the ADR. The Valori et al. study did not correlate PICI with either of these metrics. The authors note that PICI is associated with more patient and endoscopy unit variables, but variables such as level of training, years of independent practice, and trainee status are likely to be associated and therefore do not add incremental value to the use of PICI. The authors report a higher likelihood of identifying polyps in procedures with a high PICI, but evidence for PICI as a new quality indicator would be strengthened by a comparison of mean PICI to ADR for individual endoscopists.

This study raises the important subject of the patient-centered colonoscopy experience as a quality and safety indicator, but nurse-reported comfort level and sedation dose may not be the best measures for this outcome. Postprocedure patient satisfaction may be a better measure of patient experience. For monitoring of safety, 30-day postprocedure complication rate may be a better metric. Choosing cutoff values for patient comfort and sedation dose is challenging and needs further validation prior to combining these into a composite score.

The purpose of a colonoscopy performance indicator is to ensure that high quality colonoscopy is performed for each patient with the goal of minimizing missed or interval colon cancers. The authors should be congratulated for switching the focus of performance to include patient comfort, which may improve patient attitudes toward colonoscopy. However, the use of the PICI composite score for quality and safety of cecal intubation has some pitfalls, including the potential lack of applicability to other countries with different patient populations and sedation practices. Physicians need to continue to improve the quality of colonoscopy, which may incorporate the development of metrics that measure quality performance of colonoscopy, including completeness of exam, safety, patient comfort and, ultimately, the reduction of interval colon cancer.