Patient satisfaction with colonoscopy : A literature review and pilot study

1Division of Clinical Epidemiology, Research Institute of the McGill University Health Centre; 2Department of Medicine, McGill University, Montreal, Quebec Correspondence: Dr Maida J Sewitch, 687 Pine Avenue West, V-Building, The Royal Victoria Hospital, Montreal, Quebec H3A 1A1. Telephone 514-934-1934 ext 44736, fax 514-934-8293, e-mail maida.sewitch@mcgill.ca Received for publication March 31, 2008. Accepted October 10, 2008 Colorectal cancer (CRC) is the second leading cause of cancer deaths in Canada (1). CRC screening reduces the incidence of morbidity and mortality from CRC by the removal of precursor adenomatous polyps and the detection and treatment of early stage cancers, respectively (2,3). Canadian and American guidelines recommend CRC screening in persons 50 years of age and older without other identifiable risk factors for developing CRC (considered to be average risk). While there are four currently recommended screening modalities – the fecal occult blood test, flexible sigmoidoscopy, double contrast barium enema and colonoscopy – colonoscopy is considered to be optimal because it provides a view of the entire colon and an opportunity for therapeutic intervention (4-8). According to current screening guidelines, colonoscopy should be performed every 10 years after the age of 50 years (9-11). Despite universal health care in Canada, CRC screening rates are suboptimal (12,13) and more than two-thirds of people with CRC present when the disease is symptomatic (14). The use of colonoscopy as a primary CRC screening modality is controversial. In the current environment of limited financial, health care and human resources, there are ethical concerns that higher-risk individuals will be denied timely access to colonoscopy because of longer wait times. Moreover, colonoscopy does not meet the screening modality criteria for a successful screening program, which includes wide availability, low cost, ease of administration and minimal discomfort review

Patient satisfaction with colonoscopy may be related to patient compliance, because in health care, patient satisfaction is often a good predictor of patient adherence to physicianrecommended treatments or tests.In current colonoscopy practice, unsedated colonoscopy (18), long wait times that can range from 26 to 208 days (19) and inadequate information given preprocedurally (20) may negatively affect patient satisfaction.Because an unsatisfactory colonoscopy screening experience may discourage repeat screening, we reviewed studies that assessed satisfaction with the total colonoscopy experience by patients undergoing screening colonoscopy.Findings from our pilot study are also presented to understand the impact of the increased demand and limited resources for colonoscopy on patient satisfaction, and to compare levels of patient satisfaction in Canada and the United States.The findings may provide decision and policy makers with a framework for developing effective CRC screening programs.

METHODS
A MEDLINE search for articles published between January 1997 and August 2008 was conducted using the following subject headings or keywords: "mass screening" AND "colonoscopy" OR "colorectal neoplasm" AND "personal satisfaction" OR "satisfaction".Additional articles were retrieved after manual examination of the reference sections of the initial articles.

Inclusion and exclusion criteria
Prospective cohort studies were included if patient satisfaction with colonoscopy, patient willingness to return for colonoscopy under the same conditions or patient preference for colonoscopy compared with other large bowel procedures was examined.Studies were excluded if the satisfaction assessment focused on specific aspects of the colonoscopy experience (eg, pain or sedation [21][22][23][24]); if the preference assessment occurred before performing the colonoscopy procedure (25)(26)(27)(28)(29) or if interventions to improve patient satisfaction were tested (20,(30)(31)(32).Studies published before 1997 were also excluded because the colonoscopic CRC screening guidelines for average-risk individuals were not published until 1997 (33) and colonoscopies performed before this year would likely have not been for screening purposes.Furthermore, recent improvements in colonoscopes make the findings of older studies less relevant in the context of today's standard of care (34)(35)(36)(37)(38) because newer colonoscopes, which are more flexible and induce less pain compared with their predecessors (39), may increase patient satisfaction.

Quality score
The methodological quality of the studies included in the present review was scored using a grading scheme based on four criteria deemed relevant to the research question.These criteria were the following: The inclusion of persons 50 years of age and older, because • this is the recommended age at which to begin CRC screening (9-11); The method of assessing patient satisfaction was reported; • The timing of the patient satisfaction assessment was • reported; and The patient sample was restricted to those undergoing • screening procedures because less importance may be placed on satisfaction for procedures that are performed for diagnostic purposes.
Accordingly, a four-point scale was created (1 to 4), with lower scores indicating better methodological quality.Category 1 was defined as studies fulfilling the four criteria; category 2 was defined as studies fulfilling three of the four criteria; category 3 was defined as fulfilling one or two of the four criteria; and category 4 was defined as not fulfilling any of the designated criteria.

RESULTS
Based on the selection criteria, 15 studies were retained for the present review; the characteristics of these studies are summarized in Table 1.The median methodological quality score was 3 (range 1 to 4).Three studies (40)(41)(42) met category 1 criteria.Although Bosworth et al (43) did not target persons 50 years of age and older, the data were analyzed specifically for the average-risk population (category 2).One study (44) was classified as category 4 because none of the criteria were described or applicable.Although the study design variables were reported in eight studies (45)(46)(47)(48)(49)(50)(51)(52), they were classified as category 3 because the population was not restricted to persons 50 years of age and older and to those undergoing screening colonoscopy.The remaining three studies were classified as category 2; two (43,53) did not restrict their populations to those with screening colonoscopy and one study (54) targeted patients 18 years of age and older.Overall, four studies (45,46,48,52) did not report information on the indication for the procedure.Sedation was given to 100% of patients in seven studies (41,43,(48)(49)(50)53,54), and to none of the patients in three studies (44,47,51) (ie, the focus was unsedated colonoscopy).In three studies (40,42,45), the proportion of sedated patients was not detailed, and in two studies, 81% (52) and 99% (46) of patients, respectively, were sedated.
Timing of patient satisfaction in relation to undergoing the colonoscopy differed across studies.Patients underwent a colonoscopy and immediately afterwards rated their satisfaction and/or their willingness to return in eight studies (41,42,44,(46)(47)(48)51,52).In three studies (50,53,54), each participant underwent same-day CT colonography and colonoscopy, and rated satisfaction and willingness to return after each procedure and/or preference of modality several days to weeks later.In the study by Kim et al (49), patients underwent either air contrast barium enema, sigmoidoscopy or colonoscopy and rated satisfaction and willingness to return immediately after the procedure.In the study by Bosworth et al (43), all patients underwent all three procedures (air contrast barium enema, colonoscopy and CT colonography) at different times, and rated satisfaction and willingness to return immediately following each procedure; preference of modality was assessed 24 h to 72 h after the third procedure was performed.In the study by Lin et al (45), patients underwent a colonoscopy in which one-half rated their satisfaction immediately after the procedure and the other one-half during the following week.

Satisfaction with colonoscopy
Given the various measures of satisfaction used (Table 2), means and proportions of satisfied patients were reported.Three studies used rating scales with 1 being the most satisfied; one study (52) found that 96% of patients rated 1 out of 4, another found that 95% rated 1 or 2 out of 5 (50), and the third found a mean rating of 1.36±0.52(46).Using a 5-point Likert scale, Bosworth et al (43) found a mean score of 1.81, in which 1 indicated 'in total agreement with being satisfied'.Lee et al (48), using a 10 cm visual analogue scale with 10 being the most satisfied, found a mean rating of 7.2±2.6cm.Lin et al ( 45) used a 7-point rating scale with 7 being the most satisfied, and reported a mean score of 6.74±0.76.Kim et al (49) reported nonstatistically significant ORs to represent the likelihood that patients would report a higher score for one test (colonoscopy) compared with another procedure (sigmoidoscopy or air contrast barium enema) indicating similar results for the three modalities (Table 2).Seven studies (41)(42)(43)(44)47,50,52) did not report the questionnaire items, although Ristvedt et al (50) assessed levels of agreement as to how 'unpleasant' the procedure was.

Willingness to return
Nine studies (41,43,44,(46)(47)(48)(49)51,52) assessed willingness to return for the same procedure under the same conditions (Table 2).In these studies, the proportion of patients willing to return ranged from 73% to 100%.Using a 5-point Likert scale with 1 being 'in total agreement with willingness to return', one group (43) found a mean score of 1.78.Kim et al (49) found that patients undergoing colonoscopy were more willing to return than patients undergoing air contrast barium enema (OR 1.82; 95% CI 1.07 to 3.09), but not more willing than patients undergoing sigmoidoscopy (OR 1.02; 95% CI 0.66 to 1.57).

Preference of colonoscopy over another modality
Five studies compared patient preference for colonoscopy versus CT colonography (Table 2).Results were inconsistent; three studies (40,50,54) found that the majority of patients preferred CT colonography (range 58% to 72%) and two studies (43,53) found that the majority of patients preferred colonoscopy to CT colonography.

Pilot study
We conducted a prospective pilot study of patients 50 to 80 years of age about to undergo screening colonoscopy aimed at assessing patient satisfaction with the experience.Patients were enrolled on the day of the index colonoscopy and completed a mail-back questionnaire two weeks later to assess satisfaction with the colonoscopy experience using a 5-point rating scale (0 = not at all to 4 = very much).

DISCUSSION
The literature review found that patient satisfaction with colonoscopy was very high and that most patients were willing to return under the same conditions.The discrepancies among studies may have been due to the wording of the questionnaire items and/or to differences in factors that were not assessed in the questionnaires such as staff attitude, aspects of the endoscopy suite/recovery room and waiting room wait time.These other factors may be more relevant to patient willingness to return than to satisfaction with the colonoscopy itself.Preference for colonoscopy over other modalities was examined.In three studies (40,50,54), CT colonography was favoured over colonoscopy while the reverse was found in two (43,53).These equivocal findings are relevant given the recent addition of CT colonography as a recommended CRC screening procedure by some of the professional agencies in the United States (55).This newly endorsed modality would not only offer additional screening options to patients, but it would also add other professionals, equipment resources and space to the CRC screening armamentarium.Research efforts that are currently underway may inspire other organizations to include CT colonography as a recommended screening modality (40,53,56,57).Although comparison of colonoscopy with other large bowel procedures was not examined within the same individual, two studies (42,49) concluded a preference for colonoscopy compared with sigmoidoscopy, based on higher scores in the colonoscopy group.
Whereas our focus was not on specific aspects of patient satisfaction, some of the included studies reported proportions of sedated and nonsedated patients according to levels of satisfaction.Regardless of whether all (100%) patients in a study were sedated or not sedated, moderate to high levels of satisfaction and willingness to return were reported (43,(47)(48)(49)(50)(51).Moreover, in the study by Rex et al (52), in which patients were randomly assigned to receive sedation or no sedation, similar proportions of 'very satisfied' patients were found in the two study arms.Collectively, these findings suggest that sedation is not associated with patient satisfaction.
The impact of various interventions on improving patient satisfaction with colonoscopy was examined using randomized controlled trial designs.Vignally et al (32) found that a precolonoscopy consultation with a physician was associated with increased patient satisfaction.Bechtold et al (31) found that music in the colonoscopy suite was associated with increased patient satisfaction.In contrast, Bytzer and Lindeberg (30) failed to show that an informational video viewed before the endoscopy procedure was associated with increased patient satisfaction.Whereas several interventions were shown to positively affect patient satisfaction, because some may not be implementable in either a public health care system or a particular endoscopy suite due to limited resources and/or space allocation, we chose to restrict our review to observational studies.
Several strengths and limitations need to be considered in the present literature review.The major limitation of the studies reviewed is that patients who were most dissatisfied with the screening colonoscopy experience (eg, long wait times) may have gone elsewhere to have the procedure performed.However, studies included were restricted to those that assessed levels of patient satisfaction with the present standard of care; interventional studies were not included because implementing an intervention may be difficult in various settings.Higher quality scores were given to studies that included only screening populations because satisfaction is more relevant in the screening compared with the diagnostic context, in which the colonoscopy is likely being used to identify a specific problem.Although only a handful of studies were limited to screening populations, satisfaction levels were high (40)(41)(42)54) and comparable with nonscreening settings.Finally, the findings may be generalizable given that patient satisfaction was high irrespective of whether procedures occurred in a universal access, publically funded program or insurance programs requiring copayment.

CONCLUSION
After reviewing the literature regarding patient satisfaction with colonoscopy as well as our own findings, a few points are noteworthy.As a large bowel screening procedure for CRC, colonoscopy was associated with high levels of patient satisfaction and a willingness to return.Moreover, patients were equally satisfied with colonoscopy and CT colonography, a promising finding given the recent endorsement of CT colonography as a recommended CRC screening modality and the accompanying additional resources for CRC screening.Our pilot study findings are also reassuring because, even in the current era of limited resources and constraints under which colonoscopy is performed, most patients were satisfied with the care they received.Inasmuch as screening colonoscopy is advised every 10 years, providing patients with a positive colonoscopy experience is essential to encourage repeat screening.