Article Text

Download PDFPDF

Patient satisfaction with the colonoscopy procedure: endoscopists overestimate the importance of adverse physical symptoms
  1. Maaike J Denters1,
  2. Marije Deutekom2,
  3. Bert Derkx3,
  4. Patrick M Bossuyt4,
  5. Paul Fockens1,
  6. Evelien Dekker1
  1. 1Department of Gastroenterology, Academic Medical Centre, Amsterdam, The Netherlands
  2. 2Department of Social Medicine, Academic Medical Centre, Amsterdam, The Netherlands
  3. 3Department of Pediatrics, Academic Medical Centre, Amsterdam, The Netherlands
  4. 4Department of Biostatistics and Clinical Epidemiology, Academic Medical Centre, Amsterdam, The Netherlands
  1. Correspondence to Evelien Dekker, Department of Gastroenterology and Hepatology, Academic Medical Centre, University of Amsterdam, Meibergdreef 9, 1105 AZ, Amsterdam, The Netherlands; e.dekker{at}


Background Colonoscopy is a frequently performed procedure associated with a substantial burden for the patient. Most of what is known about patient satisfaction stems from surveys that target issues doctors believe to be important. It has been noticed that patients often focus on different aspects.

Aim To have patients and endoscopists rate the extent to which each of a list of patient-generated issues-of-concern contributes to patient satisfaction with the colonoscopy procedure.

Subjects A sample of consecutive patients undergoing colonoscopy in a Dutch tertiary teaching hospital and a convenience sample of endoscopists.

Methods Colonoscopy patients and endoscopists were asked to rate on a five-point Likert scale the importance of 55 items concerning the colonoscopy procedure for patient satisfaction. Items were derived from focus group sessions with colonoscopy patients. Endoscopists were invited to rate the importance of the same set of items from a patient perspective. An analysis was carried out of whether patients and endoscopists rated the importance of items differently.

Results 69 patients and 34 endoscopists completed the questionnaire. The ratings of the endoscopists were significantly different from those of patients (p<0.0001). Endoscopists underestimated the importance of involving patients in decisions, discussing risks and complications, providing the opportunity for substantive questions and offering a comfortable temperature in the examination room. Endoscopists overestimated the importance of adverse physical symptoms, such as pain and abdominal cramps, and the role of the treating doctor.

Conclusions Endoscopists do not have a good perception of the items that contribute most to patient satisfaction with the colonoscopy procedure. Overcoming this gap may be an essential step towards improving patient satisfaction by targeting those concerns most relevant to patients.

Statistics from


Colonoscopy is a frequently performed procedure. Its use is increasing, owing to an increased interest in colorectal cancer screening programmes.1 Colonoscopy is an effective diagnostic and preventive tool, but is also an invasive procedure, associated with a substantial burden for the patient.2,,8

Improving patient satisfaction with the procedure is of interest for several reasons. First, patient satisfaction is an important quality measure and increasing satisfaction with this burdensome procedure will enhance the quality of care.9 Second, improved satisfaction will probably increase compliance. A study by Bleiker et al2 showed that 25% of patients with a family history of colorectal cancer deviate from recommended screening intervals because of perceived barriers such as discomfort and embarrassment. Peña and colleagues6 showed that preprocedural nervousness was a strong predictor of unwillingness to undergo a repeat procedure. Several studies have investigated factors related to patient satisfaction with lower endoscopic procedures.3 6,,8

Most previous colonoscopy satisfaction studies were designed in a top-down manner, where researchers decided upon the items in the questionnaires. These items may not necessarily reflect patients' concerns with the procedure but merely represent what doctors believe to contribute to patient satisfaction. For example, a study investigating health concerns among children with inflammatory bowel disease (IBD) showed that paediatricians' views of what is important differed significantly from concerns stated by the children themselves.10 Several other studies have also demonstrated discrepancies between doctors' and patients' beliefs on health-related matters.11,,13 To obtain a comprehensive picture of which aspects are decisive contributors to patient satisfaction, patients should be included in suggesting items.

We aimed to explore differences in patients' and doctors' beliefs on the relative importance of patient-generated items for patient satisfaction with the colonoscopy procedure. We asked colonoscopy patients rate the extent to which they felt each of a list of 55 issues-of-concern contributed to their satisfaction with the colonoscopy procedure. In parallel, we had endoscopists rate the importance of these same issues-of-concern from a patient perspective—that is, they were invited to rate these items as they believed patients would rate them. We then evaluated differences in importance scores between patients and endoscopists and hypothesised that the importance ratings of patients would differ from the importance ratings assigned by endoscopists.


Questionnaire design

The questionnaire was designed in three steps. First, focus group interviews with colonoscopy patients who had undergone at least one colonoscopy in the past were organised to determine the items to be included in the questionnaire. We identified four separate patient groups and invited six participants from each group: patients with IBD, patients with Lynch syndrome, patients with familial adenomatous polyposis and patients who had undergone a colonoscopy in a faecal occult blood test-based colorectal cancer screening pilot. Patients were contacted for participation in the focus group sessions through our hospital and through patients associations. In the focus group sessions the participants were asked to elaborate on their experience with colonoscopy procedures and to list factors they felt were important contributors to their satisfaction with the procedure. No treating doctors were present at the focus group session, creating a safe environment. The transcripts of the focus group sessions were reviewed and an item pool of topics mentioned in the focus groups was compiled. These issues were rephrased as a list of items. This list was then presented to a number of colonoscopy experts, who made minor corrections to improve phrasing and clarity of the items. This resulted in a questionnaire of 55 items, covering the complete colonoscopy procedure, and including items on information, communication, procedural issues, physical environment, follow-up after the colonoscopy and accessibility to hospital staff.

Two versions of the final questionnaire were designed: a patient questionnaire and an endoscopist questionnaire. In the patient questionnaire, participants were asked to rate each item, expressing the extent to which they felt the specific item contributed to their satisfaction with the colonoscopy procedure. An example of the items in this questionnaire is: ‘How important is being involved in the choice for sedation to you?’. In the endoscopist version of the questionnaire, endoscopists were asked to rate each of the same set of items, expressing the extent to which they believed patients would rate the corresponding item. For this purpose, the items were rephrased. The sedation example given previously was rephrased as ‘How important do you think being involved in the choice for sedation is to patients?’. All items could be scored on a five-point Likert scale of importance, anchored at 1 (‘not at all important’) and 5 (‘very important’).

In addition to the importance items, we recorded information on patients' age and sex, their indication for colonoscopy and whether or not sedation was given. We asked endoscopists information about their age and sex, about the number of colonoscopies they had performed, the number of years they had been performing colonoscopies and the type of hospital that employed them (teaching vs regional).

Study population and setting

Study setting

The study was conducted in the endoscopy department of the Academic Medical Centre in Amsterdam, The Netherlands. The Academic Medical Centre is a tertiary teaching hospital which also serves a local community. In this hospital, an average number of 120 colonoscopies are performed each month. Before a colonoscopy, all patients are seen at the outpatient clinic where they are informed about the procedure and possible risks and complications. Colonoscopies are routinely performed under conscious sedation using fentanyl 100 μg and midazolam 5 mg. Within the endoscopy unit there is a recovery room where patients can recover from sedation after colonoscopy and before going home.


In June and November 2008, consecutive patients who had undergone a scheduled colonoscopy were invited to the questionnaire study on the day of the colonoscopy procedure. Dutch-speaking patients who were at least 18 years of age were eligible. Cognitive impairment preventing questionnaire completion was an exclusion criterion. Recruiting stopped once 120 patients had agreed to participate.


Endoscopists employed in hospitals in the Amsterdam region of The Netherlands were additionally invited to the study. Only endoscopists currently performing colonoscopies were eligible for participation.

Questionnaire administration

Patients were asked to complete the questionnaire after recovery from sedation and before leaving the hospital. Patients were considered recovered from sedation if they could carry on a normal conversation. According to Dutch guidelines no institutional review board approval was needed because the study was carried out as part of a satisfaction survey. The endoscopist version of the questionnaire was sent by post and could be returned in a postage-free envelope.

Data analysis

Mean importance scores were calculated for each item for patients and for endoscopists separately. Items were then ranked according to the mean importance score by patients. We performed a two-way repeated-measures analysis of variance to analyse whether there were systematic differences in scoring between patients and endoscopists (eg, patients gave systematically higher or lower importance scores than endoscopists) and to analyse whether importance scores differed between items (ie, items were not all scored as equally important). In addition, we looked at the interaction between groups and items, to evaluate whether importance scores differed between patients and endoscopists.

If the results of the analysis of variance showed a statistically significant interaction effect, we summarised absolute differences in mean importance scores between patients and endoscopists for each of the items. In a next substep, we identified all items that were scored differently by endoscopists and patients. Differences in mean importance scores for each item were tested for statistical significance using the Student t test statistic. In all significance tests we used a 0.05 α level to indicate statistical significance. Data were analysed using SPSS version 18.0.


Patient characteristics

Of 120 patients who underwent a scheduled colonoscopy in June and November 2008, 111 persons were eligible for participation. Seven patients had to be excluded because of an inability to understand Dutch and two patients because of cognitive impairment. A total of 69 patients (62%) completed the questionnaire. Reasons for non-participation were patient refusal (13%), logistic (72%) or unspecified (15%). Thirty-two participants (46%) were men. Mean age was 54 years. Most patients underwent a colonoscopy after referral by their general practitioner (20; 29%). Other commonly occurring indications were IBD (12; 17%) and adenoma and/or cancer surveillance (12; 17%). Thirty-six patients (52%) had had a prior colonoscopy. Sixty-six patients (96%) reported received sedation during the procedure.

Endoscopist characteristics

Fifty endoscopists were sent a questionnaire, which was returned by 34 (68%). Their mean age was 44 years. Half of the endoscopists were employed in a teaching hospital and 15 (44%) in a regional hospital. Two endoscopists did not provide information on the type of hospital in which they were employed. Participating endoscopists had been performing colonoscopies for on average 12 years (range 1–38); the mean number of colonoscopies per week was 11 (range 2–35).

Top 10 items with highest importance scores by patients

The 10 items that received the highest mean importance scores by patients are shown in table 1 (rank number in superscript behind the mean importance score by patients). Two of these items pertained to personal treatment of the endoscopist towards the patient (items 49 and 52). Four items concerned the information provision towards patients (items 3, 12, 39 and 47). Three items were about the continuity of care (items 18, 23, 44). One additional item that could not be grouped into any of the previous three categories was present in the patient top 10 (item 13).

Table 1

Mean importance scores of the complete item list by patients and by endoscopists, and significance of the difference between these mean importance scores (ranked according to magnitude of the difference)

Differences in importance scores

No structural difference was seen in response tendency between patients and endoscopist (p=0.59)—that is, patients did not give systematically higher or lower importance scores than endoscopists. As expected, not all 55 items were scored as equally important (p<0.0001). Our hypothesis that endoscopists would rate the importance of items significantly differently from patients was confirmed by a statistically significant interaction effect (p<0.0001).

Table 1 shows the mean importance score for each item by patients and by endoscopists and the difference between these mean importance scores. Items are ranked according to the magnitude of the difference (largest to smallest) and the corresponding p value of the significance of the difference is also provided. Thirteen items received significantly different important scores from patients than from endoscopists. For five of these items (1, 2, 5, 24 and 38) mean scores for patients and endoscopists were on opposite sides of the centre of the five-point importance scale. Below, we describe these differences in more detail.

Items for which endoscopists underestimated the importance in comparison with patients

Seven items received significantly lower scores from endoscopists than from patients (table 1). Three items concerned the involvement of patients in decisions surrounding the colonoscopy procedure (items 1, 2 and 13). Two additional items pertained to the information provision (items 3 and 12). Endoscopists also underestimated the importance of the feeling of uncertainty about the level of bowel cleanliness (item 10) and a comfortable temperature (item 11).

Items for which endoscopists overestimated the importance in comparison with patients

Ten items received significantly higher importance scores from endoscopists than from patients. These items were about the level of pain (item 6), the volume of bowel preparation (item 16), the abdominal pain/cramps and nausea and vomiting (items 7 and 8), the absence from work (item 5), worries about not being able to reach a bathroom (item 14) and the feeling of being in control (item 15). Endoscopists also overestimated the importance of the role of one's treating doctor (items 4, 9 and 23).


In our comparison of the views of patients and endoscopists on the relative importance of issues that contribute to a satisfactory colonoscopy procedure, we observed that endoscopists' perception differed significantly from that of patients. Endoscopists expected patients to place more emphasis on physical discomfort, such as pain during the procedure and abdominal complaints associated with the bowel preparation, while they expected less weight to be placed on the importance of involving patients in decision-making and of providing them with adequate information on, for example, possible risks and complications.

One of the strengths of this study is the close involvement of patients in the design of the questionnaire. We used input from a heterogeneous group of colonoscopy patients to determine the questionnaire content, and thus the chance of overlooking relevant concerns was minimised. A possible threat to the external validity of our study results lies in the fact that data were collected at a single centre, which was a tertiary teaching hospital. The outcomes of this study might therefore not be representative for the entire group of colonoscopy patients. However, this could only have influenced the results of the importance ratings of the items and not the item generation, since patients from multiple hospitals took part in the focus groups that were used to determine the questionnaire content.

A second limitation is that approximately one-third of the patients who underwent a colonoscopy during the study period did not complete the questionnaire. Unfortunately, we do not have data on the demographics and/or colonoscopy procedure of these patients so we cannot exclude the possibility of some form of selection bias. Since the most common reasons for non-completion appeared to be logistic issues, such as patients being in a hurry to leave the hospital, or the research assistant being busy with another patient, large differences between participants and non- participants are unlikely.

A final limitation is that patients were asked to complete the questionnaire directly after the procedure. Although we waited until they were fully recovered from sedation, this might have influenced the results. Several studies have shown that at follow-up appointments patients recall experiencing more pain than directly after the procedure.3 14 On the other hand, patients completed the questionnaire before the discussion of the results, which might have prevented influencing of their perception.

Our study is not the first to describe discrepancies in patients' and doctors' views on health-related matters. In a study by Loonen and colleagues, consecutive children with IBD attending an outpatient department were asked to rate on a Likert-scale the importance in their lives of 96 items.10 Paediatricians were offered the same set of items and were asked to do the same, but then from the child's perspective. The authors observed little overlap between items that children found most important and items that doctors thought were most important to them. As in our study, paediatricians overestimated the importance of physical symptoms while they underestimated other factors important to children (in this case worries about having a chronic disease). This study, however, was not in field of endoscopy.

Yacavone and colleagues explored factors that influenced patient satisfaction with endoscopic procedures and part of their study compared patients' and doctors' rankings of the same instrument, much in the same way as we did.8 The aim of their study was to analyse whether the modified Group Health Association of America-9 (mGHAA-9) patient satisfaction survey—a survey recommended for use in assessing satisfaction with endoscopic procedures by the American Society for Gastrointestinal Endoscopy9—covered all relevant concerns important to patients who undergo endoscopy. They added six items based on the literature to the mGHAA-9 instrument and asked both patients and doctors to rank the resulting item list from 1 to 15 in order of increasing importance. This method of prioritising items differed from our method since we asked subjects to score the relative importance of each of a list of 55 items on a Likert-scale. Yacavone and colleagues found that 11 of the 15 items were ranked significantly differently by endoscopists than by patients. Not all items that were present in their survey were present in ours and vice versa making direct comparison difficult. In accordance with our results, endoscopists underestimated the importance the doctor discussion after the procedure. In contrast to our findings, Yacavone and colleagues reported no difference in the ranking of the item that concerned adequacy of control of discomfort. In addition, several items that were not rated differently by patients and endoscopists in our study, did receive significantly different rankings in the study by Yacavone. These were the doctor's knowledge of one's medical history, waiting time before procedure, personal manner of the endoscopist and the appearance/cleanliness of the examination room. These were all ranked significantly less important by endoscopists than by patients. Finally, in accordance with our findings, no differences were seen with regard to the amount of privacy in the examination room and the adequacy of explanation of the procedure.

What causes these different views on what is important in influencing patient satisfaction? The overestimation of the importance of physical symptoms that was found in our study and also in the study assessing children's concern with IBD, might be caused because doctors have become accustomed to focusing on physical complaints and symptoms while diagnosing and treating diseases, whereas for patients the physical symptoms are just part of a larger picture that is surrounded by many uncertainties. Many of the aspects that are uncertain to patients are well known to doctors, who are used to dealing with them every day. This may be why patients want to be involved in decisions about the procedure as it might be one way of exerting some degree of control in an otherwise uncontrollable situation. Aspects of uncertainty concern patients—for instance, worries about what they are awaiting or whether or not they are going to be in any pain. Patients may value the discussion of their colonoscopy results before going home because they are worried about possible abnormalities.

Patients did not assign as much importance to the level of pain experienced as endoscopists expected, which might be because they placed more importance on items that enabled them to expect pain, such as being warned in advanced if a painful part was to be expected. Endoscopists also tended to overestimate the importance of the role of a patient's treating doctor in the colonoscopy. However, the vast majority of patients assigned a lower importance score to these items than endoscopists. This might be because endoscopists tend to remember incidents in which patients were disappointed when one was unable to perform the colonoscopy. For most patients it may be more important that the person who eventually performs the colonoscopy involves them in the process and treats them with respect. In addition, doctors in general might be more focused on the technical explanation of a medical procedure and interventions than on discussing feelings and thoughts. Future studies should explore whether interventions targeting the improvement of those concerns most relevant to patients succeed in increasing overall patient satisfaction levels with the colonoscopy procedure. Such interventions should focus on adequately informing patients about the procedure and accompanying risks and complications and on providing them with practical instructions on where to turn if they have additional questions. Furthermore, during the information consultation before the colonoscopy, patients should be actively involved in the decision-making process by discussing options for sedation, bowel preparation and timing of the procedure.

In conclusion, endoscopists do not seem to have a correct perception of which items are important contributors to patient satisfaction with the colonoscopy procedure. They overemphasise the relative importance of pain and discomfort, and underestimate the value of clear communication and shared decision-making. Improving satisfaction with the colonoscopy procedure might increase the willingness of patients to undergo the procedure and thereby the quality of care.


The authors thank Tineke Markus for her contributions to the study.



  • Contributors MJD study concept and design; acquisition of data; analysis and interpretation of data; drafting of the manuscript; statistical analysis; administration. MD: study concept and design; critical revision of the manuscript for important intellectual content. BD: study concept and design; acquisition of data; critical revision of the manuscript for important intellectual content; technical support. PB: study concept and design, analysis and interpretation of data, critical revision of the manuscript for important intellectual content, statistical analysis. PF: study concept and design, critical revision of the manuscript for important intellectual content. ED: study concept and design, study supervision; critical revision of the manuscript for important intellectual content, obtained funding.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.