The aim of this trial is to establish the position of online computer-based education as a tool for patient preparation prior to a colonoscopy. Computer based education is compared with the standard of care, nurse counseling, evaluating endoscopic quality measures and patient related outcome measures.
Improving patient education focusing on bowel preparation before a colonoscopy leads to cleaner colons. Endoscopy units must obtain informed consent and perform a risk assessment for sedative use prior to a colonoscopy. The current practice in the Netherlands to achieve these goals is nurse counseling in an outpatient setting. This is costly and has disadvantages in terms of uniformity and time consumption for both the patient and the hospital. The hypothesis is that computer-based education with use of video and 3D animations may replace nurse counseling in most cases, without losing quality of bowel cleanliness during colonoscopy.
This multicenter, randomized, endoscopist blinded clinical trial evaluates a primary outcome measure (bowel preparation) during colonoscopy. Secondary outcome measures are sickness absence, patient anxiety after instruction and prior to colonoscopy, patient satisfaction and information re-call. The study will be performed in four endoscopy units of different levels (rural, urban, and tertiary). Inclusion criteria are adult age and referral for complete colonoscopy. Exclusion criteria are Dutch illiteracy, audiovisual handicaps or mental disabilities and no (peers with) internet access.
This trial aims to establish online computer-based education as tool for patient education prior to a colonoscopy. By choosing a direct comparison with the standard of care (nurse counseling), both endoscopic quality measures and patient related outcome measures can be evaluated.
A complete colonoscopy is the procedure for detection of precancerous lesions in the colon1. For adequate examination of the colon mucosa, optimal bowel cleanliness is crucial. A poorly prepared colon leads to insufficient adenoma detection rate and therefore the need for repeated procedures. In previous studies, better patient understanding of how to prepare clearly results in a higher quality of bowel preparation2. To achieve a clean colon, patients have a restricted diet for 1-2 days and use purgatives to induce diarrhea. This elicits abdominal discomfort and interrupts daily routine. In view of these barriers, inadequate bowel preparation is not infrequent3. Optimal patient compliance to the protocol enhances effective bowel preparation and subsequent efficacy of colonoscopy.
There is appreciable variation in the way information for a colonoscopy is administered to patients4. Some patients receive information directly from their health care professional during consultation, or are informed by auxiliary personnel (nurses, technicians, or administrators), while other units provide information through printed leaflets5. The effect of any information transfer is compounded by patient dependent factors such as educational level, comprehensive capacities, and cultural aspects. This results in a mixed understanding of the information that can negatively affect compliance to instructions.
A pivotal element in patient preparation is that every patient is thoroughly informed about risks and benefits of the procedure including the bowel preparation steps for colonoscopy. In addition, the routine use of sedative and analgesics requires a risk assessment of the individual patient. Many centers rely on nurse counseling to obtain informed consent before the procedure. This results in patient improved adherence to the instructions for bowel preparation. However, while effective, it is time-consuming for the nurse, repetitive, and results in patient-to-patient variability of information. More importantly, it demands an extra hospital visit for the patient, implicating absence of the patient at work6. In summary, it is an economically challenging practice in cost-conscious healthcare environments. Previous studies show that focused e-learning paths enable good comprehension and learning and enhance patients satisfaction7. Web-based education is used successfully for increasing knowledge of patients and it has become an accepted mechanism for obtaining informed consent. This has led to the development of tailored instruction programs for bowel preparation that combines the advantages of flexibility in time and environment, yet maintains consistency in delivery of information. Previously, the authors developed a tool that allows computer assisted instruction (CAI) for colonoscopy8. This tool employs a computer animation that captures the viewers’ attention while adequately informing him/her of objectives for colonoscopy. Written in comprehensible language in logical order, the module educates patients on different aspects of colonoscopy. It provides basic anatomical teaching points and step-by-step instructs the patient how to perform bowel preparation. In our pilot study we showed that CAI for colonoscopy enhanced bowel preparation to the level that is comparable to nurse counseling.
The research group sought to enhance the efficacy of the developed CAI. Its limitation was that it was a unidirectional tool that delivered information but did not allow acquiring patient specific information concerning medical history and medication use. This is an important part of the nurse counseling visit, as it allows a pre-sedation risk assessment when judged by the nurse. Therefore, a dedicated questionnaire was developed, designed to collect data points for structured risk assessment. This questionnaire is completed by the patient at the end of the CAI. This eliminates the need for a face-to-face meeting with a nurse or physician at this point in time. The use of two-way communication (combining CAI with a questionnaire) is practical and provides high quality information to the patient whilst at the same time attending to the need of the endoscopist for information on sedation risks. This combined instruction and acquiring of information is known as computer based education (CBE)7.
The goal of this trial is to test the utility, practicality, and patient-perceived usefulness of CBE off-center, in comparison to conventional nurse counseling. The hypothesis is that CBE is non-inferior to nurse counseling in achieving high quality of bowel preparation during colonoscopy. This process is independent of time and space and therefore can be viewed in the comfort of the patients’ home. Accordingly, the chosen secondary outcomes are patient related outcome measures such as a short leave absence, anxiety, satisfaction and comprehension of information, as these might benefit from delivery through this digital channel. Included process measures are patient activation, health and e-health literacy to determine which patients benefit most from this tool.
Study design
The trial is set up as an endoscopist blinded multicenter randomized controlled trial design. Inclusion criteria are adult age and a referral for elective complete colonoscopy. Exclusion criteria are illiteracy in Dutch and significant audiovisual handicaps and mental disabilities that preclude delivery of CBE. Also, patients were excluded if there is no internet access or a relative with internet access (see Table 1). Patients will be recruited by back office staff at the outpatient’s clinic in 4 large volume endoscopy centers in the Netherlands. All patients receive a split dose laxative regime based on either polyethylene glycol or sodium picosulfate. After evaluation of in- and exclusion criteria by trained staff, patients are randomized in 1:1 distribution per trial site using a randomization tool (described in the protocol below). Reasons for declining to participate are recorded. The trial flowchart is presented in Figure 2.
Outcome measures
The primary outcome measure is the quality of bowel preparation during endoscopy. Endoscopists are trained to score the bowel preparation with the Boston Bowel Preparation Scale (BBPS). The BBPS is a cumulative score of three bowel segments, ranging from 0-1 “unsatisfactory”, 2-3 “poor”, 4-5 “fair”, 6-7 “good”, and 8-9 “excellent”. Scores of ≥6 are considered adequate9,10. As secondary outcomes, the focus is on sickness absence, anxiety, satisfaction and information re-call. Information is also collected on patient activation and health literacy.
The cost minimization effect of the intervention is calculated in two ways. The comparison between groups with regard to endoscopy unit costs will be done using a cost-per-visit analysis. The macroeconomic effect of sickness absence is also evaluated, as patients in the intervention group will need less hospital visits. To do so, several items are assessed: socio-economic status, work status and duration of sickness absence, using an adapted iProductivity Cost Questionnaire11.
Patients anticipating invasive medical procedures often experience anxiety that may exceed their coping mechanisms. Anxiety is assessed at T0 and T1 with the State-Trait Anxiety Inventory (STAI)12. The STAI is a widely used 20-item self-report instrument with scores ranging from 20 (absence of anxiety) to 80 (high anxiety). Patient satisfaction is scored using two different measures. Patient experience impacts future behavior and therefore “willingness to return” is assessed at T3, ranging from 1 (extremely unwilling to return) to 10 (extremely willing to return). Furthermore, the Net Promoter Score (NPS) is utilized on the question “Would you recommend this endoscopy unit to your peers?”. Patient’s scores range from 1 (Not at all likely) to 10 (Extremely likely). The NPS will be assessed at T0 and T3 and is calculated as % Promoters (scores 9-10) – % Detractors (scores 1-6)13. To evaluate patient comprehension of the information in the CBE patients are asked to reproduce elements of the instruction. The patient information re-call is assessed at T1 (before colonoscopy) using a 10-item test, with questions to be answered with “yes” or “no”. The effect of patient education in colonoscopy is influenced by the patient ability to understand medical information. The 14-item Dutch validated Health Literacy Scale is used to assess this item, divided in 3 subscales, at T014. A new 21-item questionnaire is added as a measure for e-Health Literacy15. This contains questions regarding the skill and experience of patients in handling medical information online. Patients are confronted with options every day that may have major implications for their health. Effectively managing their choices requires knowledge, skill, and confidence. To this end these elements were mapped at T0 13-item Patient Activation Measure Scale (PAM-13)16. The current health status of patients is evaluated with the Medical Outcomes Study 36-item health survey (RAND-36) at T017.
Statistical analysis
To statistically compare both groups on the primary outcome, the relative risk for an inadequately prepared colon, defined as a BBPS <6, is used. In literature, a 90% success rate (for an adequately prepared colon) is common, with a 10% non-inferiority margin as the maximum clinically acceptable difference. The non-inferiority power calculation resulted in 180 patients per group, 360 patients in total. This is required to exclude a difference in favor of the standard group of more than 10%. With a margin of ± 60% loss of patients before completing the protocol, based on earlier research, the target number of patients to approach is set at 1,000. In addition to the non-inferiority analyses, superiority analyses will be conducted to investigate effects on secondary outcome measures.
The study is authorized by the ethics review board of the Radboud University Medical Center (#2015-1742). Subsequent approval of the executive boards from each of the participating institutes is obtained (Trial registration: Dutch Trial Registry, NTR 5475).
1. Enrolling patients in the trial/randomization
2. Baseline questionnaire
3. Intervention arm: patient is prepared with computer-based education
4. Control arm: Patient is visiting the outpatient clinic
5. Day of colonoscopy
The earlier mentioned pilot study compared nurse instruction to CAI using the same interactive tool as used in this protocol8. As the goals of this study were comparable to the outcomes used in this protocol, a short explanation of the results of the pilot are provided here in more detail. See also Table 28.
In this pilot study 385 patients were enrolled. The CAI group contained 188 subjects. The control group receiving nurse counseling had 197 patients. The baseline characteristics were evenly distributed between CAI and nurse counseling. No significant differences were found comparing groups on bowel preparation scores, using two different scales. In the BBPS analysis nurse vs. CAI group scores were adequate: 6.54 ±1.69 vs. 6.42 ±1.62. In the Ottawa Bowel Preparation Scale, scores were 6.07 ±2.53 vs. 5.80 ±2.90 respectively. On secondary measures, the enquired patient comfort was significantly higher in the CAI group shortly before colonoscopy. Aar five-point Likert scale was used, ranging from 1 (low) to 5 (high). Results were 4.29, ± 0.62 in the CAI group vs. 4.42, ± 0.68 in the nurse counseling group. As this rating was higher directly after nurse counseling, there is influence of the human factor for personal contact and offering emotional support. Anxiety and information re-call scores showed no statistical difference (see Table 38).
Figure 1. An overview of the computer based education before colonoscopy used in this trial, illustrating all the steps in the patients’ journey. The lower right screen depicts the questionnaire for pre-sedation risk assessment and written informed consent. Please click here to view a larger version of this figure.
Figure 2. Flowchart trial with time points Please click here to view a larger version of this figure.
Inclusion criteria | Exclusion criteria |
Adult age | Illiteracy for Dutch |
Referral for complete colonoscopy requiring bowel preparation | Audiovisual handicaps |
Able to provide informed consent | Mental disabilities |
Unwilling to participate | |
No internet access (or relatives with internet access) |
Table 1. In- and exclusion criteria
Nurse counseling | Computer Assisted Instruction | Nurse versus Computer Assisted Instruction (Mann-Whitney) | |
(n, % scoring rate) | (n, % scoring rate) | ||
Comfort Score after consult/CAI (T1) (1=very low, 5=very high) |
Mean 4.54, ± 0.56 | Mean 4.17, ± 0.51 | p = 0.000 |
(n=193, 98.0%) | (n=188, 100%) | ||
Comfort Score before endoscopy (T2) (1=very low, 5=very high) |
Mean 4.29, ± 0.62 | Mean 4.42, ± 0.68 | p = 0.039 |
(n=162, 82.2%) | (n=124, 66.0%) | ||
Comfort Score after endoscopy (T3) (1=very low, 5=very high) |
Mean 4.16, ± 0.93 | Mean 4.28, ± 0.84 | P = 0.322 |
(n=150, 76.1%) | (n=117, 62.2%) | ||
Anxiety Score after consult/CAI (T1) (5=very low, 1=very high) |
Mean 3.16, ± 1.30 | Mean 2.92, ± 1.22 | p = 0.071 |
(n=193, 98.0%) | (n=188, 100%) | ||
Anxiety Score before endoscopy (T2) (5=very low, 1=very high) |
Mean 2.80, ± 1.32 (n = 162, 82.2 %) | Mean 2.90, ± 1.27 (n = 124, 66.0%) | p = 0.451 |
(n=162, 82.2%) | (n=124, 66.0%) | ||
Knowledge and Comprehension 10 item test score before endoscopy | Mean 7.08, ± 1.17 (n = 164, 83.2 %) | Mean 7.31, ± 1.11 (n = 127, 67.6%) | p = 0.112 |
(n=164, 83.2%) | (n=127, 67.6%) |
Table 2. Bowel preparation scores in our earlier pilot study8
Nurse counseling | Computer Assisted Instruction | Nurse versus Computer Assisted Instruction (Mann-Whitney) | |
(n, % scoring rate) | (n, % scoring rate) | ||
Ottawa Bowel Preparation Scale (mean, SD) | 6.07, ±2.53 | 5.80, ±2.90 | p = 0.418 |
(n=115, 58.4%) | (n=87, 46.3%) | ||
Boston Bowel Preparation Scale (mean, SD) | 6.54, ±1.69 | 6.42, ±1.62 | p = 0.576 |
(n=129, 65.5%) | (n=88, 46.8%) |
Table 3. Secondary outcomes in our earlier pilot study8
Supplemental video: An instructive video on how the computer-based education is implemented in the endoscopy unit can be found here: https://vimeo.com/141342029
The E-Patient Counseling (E-PACO) trial aims to study the utility, practicality, and patient-perceived usefulness of computer-based education (CBE), in comparison to conventional nurse counseling. In this manuscript the CBE is demonstrated together with the methodology used to evaluate the hypotheses.
It is established that high quality colonoscopy is the golden standard for prevention of colorectal cancer. Inadequate bowel preparation is related to the missing of neoplasm’s and increase need for repeat examinations with increased costs and cumulative discomfort for patients18,19,20. The cleanliness of the colon or bowel preparation is the main quality measure and therefore used as primary outcome measure. Studies that focus on patient education prior to colonoscopy have yield significantly better results in bowel cleanliness for their intervention (cartoons, day-before-colonoscopy reminder calls and nurse counseling)21,22,23. However, some of these trials are derived from non-Western populations, so cultural differences might hinder generalizing these findings in Western population.
The pilot study did not find significant differences, so a non-inferiority design is chosen. If this intervention proves to be non-inferior, the operational advantages of counseling at home (reducing personnel and facility costs) still outweigh the investment for endoscopy units. There might be potential gain in the patient related outcome measures like anxiety and satisfaction. For generalization purposes it is of great importance to acquire a large heterogeneous sample that is representative for all patients in a (Western) endoscopy unit. By using four endoscopy units in several Dutch provinces (based in rural, urban and academic hospitals) the aim is to optimize diversity.
Possible influences educating patients are health literacy, educational level and the time between education and the procedure. When the intervention was designed, the perspective from patient panels, nurses and doctors were all incorporated. Lessons learned in other best practices, such as 3D visualization, were implemented. This takes into account the possibility of variation in learning styles between individuals and increases the potential for acquisition and retention of knowledge. The use of voice-over in adjunction to video accommodates patients with low literacy levels. From the elderly user perspective, easily accessible program features are added, such as optionally enlarged fonts and utilizing touch screen. Unlimited access to the information is guaranteed though a re-usable web-based link, so patients are enabled to view their CBE on-demand. Finally, language barriers are easily overcome with the availability in the menu to choose the language.
The double-check of information derived from the questionnaire also reinforces patients to important constructs of information provided earlier. Although guided by logical transitions at first time viewing, user control over the program sequence for repeated learning is allowed. Before the implementation, there was a careful analysis performed to provide a seamless integration of the CBE in the current endoscopy unit process.
A multicenter trial in real life setting has barriers for inclusion. For the clinically gathered questionnaires the usual contact moments were chosen to hand out questionnaires by the endoscopy unit operational staff. Missing questionnaires can be the result. Nevertheless, this trial aims to collect all relevant information at all time points.
Patients are eligible for the trial and can operate the CBE even with very basic computer skills. But in the lowest literacy category, it is not possible to test the hypotheses. As of this, it is important to maintain the possibility of face-to-face patient education in the route towards the endoscopy suite for this group.
As the future will provide more challenges in patient education, more research in this field is important. The method presented is suitable for evaluating the use of CBE in other endoscopic procedures, as well as in other departments.
The authors have nothing to disclose.
The authors thank Dr. Wietske Kievit for her methodological support and comments on the manuscript.
Computer Based Education | Medify BV | n/a | Computer Based Education tool for patient instruction prior to colonoscopy |
Computer / tablet | Any | A computer or tablet should be used to complete the e-learning and fill out the questionnaires | |
Medify Content Management System | Medify BV | n/a | A content management system to process the e-learning content |