Colorectal Cancer: Cost-effectiveness of Colonoscopy versus CT Colonography Screening with Participation Rates and Costs
Abstract
Purpose
To compare the cost-effectiveness of computed tomographic (CT) colonography and colonoscopy screening by using data on unit costs and participation rates from a randomized controlled screening trial in a dedicated screening setting.
Materials and Methods
Observed participation rates and screening costs from the Colonoscopy or Colonography for Screening, or COCOS, trial were used in a microsimulation model to estimate costs and quality-adjusted life-years (QALYs) gained with colonoscopy and CT colonography screening. For both tests, the authors determined optimal age range and screening interval combinations assuming a 100% participation rate. Assuming observed participation for these combinations, the cost-effectiveness of both tests was compared. Extracolonic findings were not included because long-term follow-up data are lacking.
Results
The participation rates for colonoscopy and CT colonography were 21.5% (1276 of 5924 invitees) and 33.6% (982 of 2920 invitees), respectively. Colonoscopy was more cost-effective in the screening strategies with one or two lifetime screenings, whereas CT colonography was more cost-effective in strategies with more lifetime screenings. CT colonography was the preferred test for willingness-to-pay-thresholds of €3200 per QALY gained and higher, which is lower than the Dutch willingness-to-pay threshold of €20 000. With equal participation, colonoscopy was the preferred test independent of willingness-to-pay thresholds. The findings were robust for most of the sensitivity analyses, except with regard to relative screening costs and subsequent participation.
Conclusion
Because of the higher participation rates, CT colonography screening for colorectal cancer is more cost-effective than colonoscopy screening. The implementation of CT colonography screening requires previous satisfactory resolution to the question as to how best to deal with extracolonic findings.
© RSNA, 2018
Introduction
Colorectal cancer (CRC) is the second most common cause of cancer mortality in the Western world (1). Early detection and treatment of CRC can reduce CRC incidence and mortality. Several randomized controlled trials have demonstrated a mortality reduction with use of guiac fecal occult blood tests (2–5) and sigmoidoscopy (6).
Colonoscopy is considered the reference standard and allows for direct removal of colonic lesions. There is only indirect evidence of the preventive effect, however, and its use as a primary screening test has drawbacks, including its perceived burden, low rates of patient compliance with screening recommendations, risk of complications, costs, and need for endoscopy capacity (7). This burden can be reduced by selecting individuals at increased risk for CRC to undergo colonoscopy, for instance by incorporating a less-invasive primary test and offering colonoscopy only to those with positive results. Computed tomographic (CT) colonography is an example of such a test, with the benefit that it is a less-invasive screening method, needs less-extensive bowel preparation (8), and has a lower risk of complications (9).
In several studies, the accuracy and one-round yield of advanced neoplasia was slightly higher with colonoscopy than with CT colonography (10–12). In addition, most of the comparative cost-effectiveness studies showed that colonoscopy screening strategies were more cost-effective than CT colonography strategies when assuming full participation (13–15). However, the outcomes of these analyses were highly sensitive to the participation rates and costs of the two screening tests.
With a lower participation rate, screening is applied in fewer individuals, decreasing both the effectiveness and costs. Therefore, differences in participation between tests will influence their relative cost-effectiveness. In 2012, a large randomized controlled trial in the Netherlands (the Colonoscopy or Colonography for Screening [COCOS] trial) was the first to compare the participation, yield, and unit costs of CT colonography and colonoscopy in a dedicated screening setting (12). Participation was indeed higher for CT colonography (33.6%, 982 of 2920 invitees) than for colonoscopy (21.5%, 1276 of 5924 invitees) (12), whereas unit costs for CT colonography were 25% lower than those of a negative colonoscopy and 53% lower than those of a positive colonoscopy (because of additional costs for colonoscopy with polypectomy) (16,17). These estimates allowed for a more representative cost-effectiveness analysis of CT colonography and colonoscopy screening. Therefore, the aim of this study was to determine the comparative cost-effectiveness of CT colonography versus colonoscopy screening with use of data on unit costs and participation rates from a randomized controlled screening trial in a dedicated screening setting (at the time of publication, the current conversion rate to U.S. dollars was 1.18).
Materials and Methods
The COCOS Trial
Costs and participation data were obtained from the COCOS trial, a randomized controlled trial of CT colonography versus colonoscopy screening (12,16,17). In this trial, screening-naive members of the general population aged 50–75 years and living in the regions of Amsterdam and Rotterdam were identified by means of the regional municipal administration registries and randomly allocated (ratio, 2:1) to invitation to primary screening with colonoscopy or CT colonography. The participation rate in the first screening round was 21.5% (1276 of 5924 invitees) for colonoscopy versus 33.6% (982 of 2920 invitees) for CT colonography (12). A positive finding at CT colonography was estimated to cost €158, a negative finding at CT colonography €149, a positive finding at colonoscopy (including polypectomy and pathology costs) €329, and a negative finding at colonoscopy €192 (16,17). Details on bowel preparation, protocols, and yield are given in Appendix E1 (online).
Microsimulation Model
We used a previously described microsimulation model for CRC (18,19) (Appendix E1 [online]). A microsimulation model can incorporate trial results to estimate the lifetime effects of findings of the randomized trial. In addition, the model can estimate the cost-effectiveness of various screening strategies instead of only the screening strategy used in the randomized trial and thereby study which screening strategies are optimal.
This model simulates the life histories of a large population of individuals from birth to death. As a simulated person ages, one or more adenomas may develop and may progress in size from small (≤5 mm) to medium (6–9 mm) to large (≥10 mm). Some adenomas can develop into preclinical cancer, which may progress through stages I to IV. At any time, the process may be interrupted by death from another cause.
Screening may prevent the development of cancers because it enables detection and removal of adenomas and it may help detect CRC in an earlier stage with a more favorable survival. Thus, CRC incidence and mortality could be reduced. The life-years gained with screening are calculated as the difference in model-predicted life-years lived in the population with and population without CRC screening. The natural history is further described in Appendix E1 (online).
Study population.—In this study, we modeled the age distribution of the Dutch population aged 25–85 years in 2015 (20) and followed up all individuals until death. Life expectancy was based on sex-specific life tables from 2011 obtained from Statistics Netherlands (20).
Screening strategies.—Screening started in 2015. We simulated four different strategies for referral of CT colonography–positive individuals to diagnostic colonoscopy (Fig 1). Individuals were offered different screening schedules varying according to (a) age to start screening (40, 45, 50, 55, 60, and 65 years), (b) age to stop screening (70, 75, 80, and 85 years), and screening interval (3, 5, 10, 15, and 20 years).
Figure 1: Diagram shows the four simulated follow-up strategies for CT colonography (CTC ) screening that differ when 6–9-mm adenomas are found. In all four strategies, individuals with lesions of at least 10 mm at CT colonography were immediately referred for diagnostic colonoscopy, and individuals without lesions or with lesions measuring 1–5 mm at CT colonography returned to the screening program. Follow-up strategies differed in management of medium-sized lesions (6–9 mm) seen at CT colonography. Individuals with medium-sized lesions (1) were directly referred for diagnostic colonoscopy (ie, by using a cutoff of 6 mm); (2) were returned to screening program (ie, corresponding with a 10-mm cutoff); (3, 4) were offered follow-up CT colonography after 3 years, as was done in COCOS trial, and referred to diagnostic colonoscopy if they at follow-up CT colonography had a medium or large lesion (3) or a large lesion (4). In the 4th strategy, persons with medium-sized adenomas continued to receive follow-up CT colonography, either until a large lesion was detected or until a medium lesion was no longer detected.
Together, the various screening ages and intervals resulted in 86 screening strategies for colonoscopy and each follow-up CT colonography strategy, combined for a total of 480 strategies.
If adenomas were detected during primary or diagnostic colonoscopy, they were removed and the individual entered surveillance according to the Dutch guidelines (21).
Test characteristics and size distribution of adenomas.—We assumed the sensitivity of colonoscopy to be 75% for adenomas with a diameter of 1–5 mm, 85% for adenomas measuring 6–9 mm, and 95% for adenomas measuring at least 10 mm and CRC (22). The specificity of colonoscopy was assumed to be 90% to account for the presence of nonadenomatous polyps. We assumed the sensitivity of CT colonography to be 75.7% for adenomas measuring 6–9 mm, 85.9% for adenomas of at least 10 mm, and 95% for colorectal carcinoma. Specificity was assumed to be 91.4% with a cutoff of 6 mm and 97.6% with a cutoff of 10 mm (10,23). We validated the model with these test characteristics with observations in the COCOS trial (Table E1 [online]) and adjusted the size distribution of adenomas (but not the prevalence) in the model to fit the data best.
Participation.—We first assumed a 100% participation rate to identify the optimal screening strategies in terms of screening interval, age range, and CT colonography cutoff. Subsequently, we simulated the optimal screening strategies with the observed participation rates to compare the two screening tests in terms of cost-effectiveness.
We assumed 100% participation to determine the optimal strategies because assuming observed participation could result in optimal screening strategies for the population that are at too-short intervals. These short intervals in the population will result in average screening intervals at the most optimal length for an individual. However, this would lead to overscreening in those who adhere to recommendations, which we believe is unethical and which in turn might also lead to lower compliance in practice than assumed in the analysis.
To compare the two screening tests in terms of cost-effectiveness, we then simulated the optimal screening strategies with the observed participation rates (21.5% [1276 of 5924 invitees] for colonoscopy vs 33.6% [982 of 2920 invitees] for CT colonography). The observed age dependency on and participation in diagnostic colonoscopy after positive CT colonography were modeled accordingly. Participation with surveillance colonoscopy was assumed to be 80% (24). In a sensitivity analysis, we also modeled all possible screening strategies with observed participation.
Data on subsequent participation in colonoscopy or CT colonography screening are lacking. We assumed stable overall participation in subsequent rounds and assumed that 90% of the previous responders also attended the subsequent screening round, as found in a Dutch fecal immunochemical test (FIT) screening trial (25). To maintain stable overall participation, the remaining percentage was filled with previous nonattenders. In addition, we assumed that 10% of the individuals never attended screening (24) and that the risk of CRC in this group was higher than that in the general population (relative risk = 1.15) (2).
Costs and utility losses.—We included all costs from a third-party payer perspective. We assumed unit costs for CT colonography and colonoscopy as observed in the COCOS trial. We further included screening and treatment costs and utility losses as presented in Table 1.
 |
Extracolonic findings.—Extracolonic findings were not included in the cost-effectiveness analysis because long-term follow-up data are lacking. We did, however, include a sensitivity analysis accounting for the currently available data (Fig E6b [online]). Two scenarios were simulated. In the first scenario, only extra costs and utility loss due to extracolonic findings were accounted for. In the second scenario, we also assumed a benefit in the detection and further management of abdominal aortic aneurysms.
Outcomes
We used the model to predict costs and quality-adjusted life-years (QALYs) gained for all screening strategies compared with no screening. Costs and QALYs gained were discounted by 3% per year in accordance with the international literature (32). Alternatively, Dutch discounting (1.5% for QALYs and 4% for costs) was used in a sensitivity analysis.
Analysis
We first showed disaggregated outcomes behind the total costs and QALYs gained finally used in the cost-effectiveness analysis of screening every 10 years from age 50 to 70 years for colonoscopy and CT colonography screening with a 6-mm cutoff.
Next, the total costs and QALYs gained of all 480 simulated strategies were compared. Per test, it was determined which strategies were efficient by ruling out strategies that were more costly and less effective than other strategies (simple dominance) or combinations of other strategies (extended dominance). On a plot of QALYs gained versus costs, the line that connects the efficient strategies is called the efficient frontier. The incremental cost-effectiveness ratio (ICER) of each efficient strategy was calculated by comparing its costs and effects with those of the next less-costly and less-effective efficient strategy. An ICER of less than €20 000 per QALY gained was assumed to be cost-effective (33).
Sensitivity Analysis
In addition to the above-mentioned sensitivity analysis, we performed one-way sensitivity analyses on several other parameters, as summarized in Table 1.
Results
Screening Every 10 Years
With a 100% participation rate, screening once every 10 years from age 50 to 70 years resulted in a higher mortality reduction and more QALYs gained with colonoscopy than with CT colonography (106 vs 81 QALYs gained, 24% lower) (Table 2). With colonoscopy screening, the screening and surveillance costs were higher than with CT colonography, but the CRC treatment savings were also higher. This resulted in lower total costs of the colonoscopy screening program. Therefore, with this screening strategy, colonoscopy dominated CT colonography screening.
 |
With observed participation, CT colonography screening resulted in a higher mortality reduction, more QALYs gained (29 vs 22 QALYs gained, 34% higher), but still higher total costs. The number of lifetime colonoscopies and complications and the number of people needed to scope to detect one advanced neoplasia were lower with CT colonography screening, assuming both 100% and observed participation.
Optimization of Screening Interval, Age Range, and CT Colonography Cutoff
When all strategies were considered with a 100% participation rate (varying screening age range and interval), only colonoscopy strategies appeared on the efficient frontier (Fig 2). The optimal colonoscopy and CT colonography strategies are presented in Table 3. The most effective colonoscopy strategy with an ICER below the Dutch threshold of €20 000 per QALY gained was colonoscopy every 5 years from age 50 to 70 years (strategy 4 in Table 3).
Figure 2: Graph shows modeled costs and quality-adjusted life-years (QALYs) gained per 1000 participants for four CT colonography (CTC) screening strategies and colonoscopy screening, with different starting and stopping ages and screening intervals, with 3% discounting, from a participant’s perspective. * = Strategies on this frontier are presented in Table 3. † = CT colonography strategy number corresponds with numbers presented in Figure 1.
 |
Of the CT colonography strategies, those with a 6-mm cutoff resulted in more QALYs gained and lower total costs than other follow-up strategies. The most effective CT colonography strategy with an ICER below the Dutch threshold of €20 000 per QALY gained was CT colonography with a 6-mm cutoff every 3 years from age 45 to 80 years (strategy 8 in Table 3).
Comparison of CT Colonography and Colonoscopy
When the optimal colonoscopy and CT colonography strategies were simulated with observed participation, both costs and QALYs gained decreased compared with full participation (Fig 3, Table 3). The costs and QALYs gained assuming observed participation rates decreased more for colonoscopy screening than for CT colonography screening. Still, colonoscopy screening with one or two lifetime screenings was less costly than and just as effective as the same CT colonography strategies. However, with more lifetime screening, CT colonography screening dominated the colonoscopy strategies. The first CT colonography strategy on the efficient frontier had a screening age of 55–70 years, an interval of 5 years, and an ICER of €3162 per QALY gained. The most effective CT colonography strategy with an ICER below the €20 000 threshold was CT colonography triennially from age 45 to 80 years (ICER €14 709 per QALY gained).
Figure 3: Graph shows modeled costs and quality-adjusted life-years (QALYs) gained per 1000 invitees of CT colonography (CTC) strategies with 6-mm cutoff and colonoscopy with different starting and stopping ages and screening intervals, with 3% discounting, from a population’s perspective. * = CT colonography strategy 1 in Figure 1.
Sensitivity Analyses
In concordance with the base case analyses, most sensitivity analyses showed that colonoscopy with one or two lifetime screenings dominated CT colonography screening, whereas, with more lifetime screenings, CT colonography dominated the colonoscopy strategies (Fig E6 [online]). Simulating all screening strategies with observed participation did not change which screening strategies were efficient and produced the same pattern (Fig E6a [online]). Including extracolonic findings made CT colonography less cost-effective in the version without benefits of detected abdominal aortic aneurysms and more cost-effective in the version with benefits of detected abdominal aortic aneurysms, whereas CT colonography still dominated colonoscopy screening in both cases, showing an ICER of €3478 and €2458 per QALY gained, respectively (Fig E6b [online]). If CT colonography costs were doubled or colonoscopy costs halved, colonoscopy dominated CT colonography and vice versa (Fig E6c, graphs I–N [online]). In addition, when subsequent participation was random, colonoscopy dominated CT colonography screening (Fig E6c, graph V [online]).
Discussion
The results of this study confirm that, for people willing to participate in CRC screening, colonoscopy is more cost-effective than CT colonography screening. However, from a population’s perspective, with participation as observed in the COCOS trial, colonoscopy screening is less cost-effective than CT colonography screening if the latter offers more than two lifetime screenings. Because the ICER of the least-intensive CT colonography strategy on the efficient frontier was well below the Dutch threshold of €20 000 per QALY gained, CT colonography screening is preferred over colonoscopy screening as a one-test–based screening program on a national level.
With observed participation and one lifetime screening, the lower sensitivity of CT colonography screening was compensated for by the higher participation rate, resulting in the same number of QALYs gained for CT colonography and colonoscopy. However, the costs of the diagnostic colonoscopy added to the CT colonography screening costs made CT colonography more expensive. The dominance of CT colonography with more lifetime screenings can be explained by the yield of the additional screening rounds. After each screening round, the prevalence of (advanced) neoplasia in the screened population decreases. Because sensitivity for adenomas is higher for colonoscopy than for CT colonography, the residual number of adenomas is higher in patients undergoing CT colonography screening than in those undergoing colonoscopy screening. Owing to the higher participation rate, the number of patients undergoing CT colonography screening is also higher than the number undergoing colonoscopy screening, further increasing the cost-effectiveness of intensifying screening in CT colonography compared with colonoscopy.
Previous studies assuming a 25% higher participation rate for CT colonography estimated that the costs of CT colonography should be no higher than 75%–95% of a colonoscopy to be more cost-efficient than colonoscopy (34,35). In our analysis, participation in CT colonography was 56% higher than that with colonoscopy (33.6% vs 21.5%, respectively), whereas the costs of a negative and positive CT colonography were 75% (€149 of €192) and 47% (€158 of €329) that of a negative and a positive colonoscopy, respectively. Thus, the relative participation rate of CT colonography was higher and relative costs of CT colonography lower than the studied threshold; therefore, our results are in line with results of these studies.
Because earlier cost-effectiveness analysis showed that outcomes were highly sensitive to assumed participation rate and test costs (13–15), an important strength of our study is that it is, to our knowledge, the first cost-effectiveness analysis that uses real-life data on unit costs and participation in a similar setting.
Four limitations of the study should be addressed. Lacking data on participation to subsequent screening rounds, we assumed participation for subsequent rounds was stable and that 90% of previous responders also participated in a subsequent round (25,36–38). For the first aspect, although at this moment we have no concrete indication that participation will improve quickly, we acknowledge that with increased promotion, participation of colonoscopy and CT colonography might improve over the years and participation of colonoscopy may then come close to that of CT colonography screening. The participation rate of colonoscopy should be quite close to that of CT colonography to dominate CT colonography screening (the rate for CT colonography should decrease to 22.5%). For the latter aspect, we explored two alternative scenarios in the sensitivity analysis, of which the outcome was that colonoscopy screening is preferred when assuming random subsequent participation, a scenario we find highly unlikely.
Second, we did not include two relevant aspects: exposure to ionizing radiation and extracolonic findings at CT colonography. With regard to radiation, Berrington de Gonzalez et al (39) estimated the ratio of CRCs prevented to the number of radiation-related cancers induced at 24:1 to 35:1. The ionizing radiation dose used in that study was substantially higher than that used in the COCOS trial and in the studies on which our CT colonography sensitivity was based. With respect to extracolonic findings, we conducted a sensitivity analysis with the sparse data on costs and abdominal aortic aneurysm screening, which showed a decrease in cost-effectiveness of CT colonography in the version without benefit of detected abdominal aortic aneurysm and an increase in cost-effectiveness in the version with benefit of detected abdominal aortic aneurysm. As long-term follow-up of people with other screening-detected extracolonic lesions is lacking, the U.S. Preventive Services Task Force concluded that the effect of extracolonic findings on the cost-effectiveness of CT colonography screening as yet cannot be estimated (40). In theory, including extracolonic lesions could make CT colonography screening more cost-effective. On the other hand, detection of extracolonic lesions could be harmful, thereby reducing the (cost) effectiveness of CT colonography screening or even making CT colonography screening harmful in general. One could consequently argue that clinicians should ignore extracolonic findings, which could lead to other unacceptable ethical dilemmas. As long as this dilemma is unresolved, it is unknown whether CT colonography screening fulfills the World Health Organization criterion that overall benefit should outweigh the harms (41).
Third, we did not explicitly model distinct pathways for traditional and sessile serrated adenomas and/or polyps. The average time it takes for an adenoma to develop into CRC was calibrated to the randomized U.K. Flexible Sigmoidoscopy Screening Trial (42) and included both traditional adenomas and sessile serrated adenomas and/or polyps. Therefore, both adenoma types are included in the modeled mix of slowly and rapidly progressing lesions. An explicit separate pathway would be relevant if indeed CT colonography is less sensitive than colonoscopy for sessile serrated adenomas and/or polyps (because they are often flat and therefore harder to detect with imaging [43]) and if the malignant potential of sessile serrated adenomas and/or polyps toward cancer is different from the traditional pathway, which remains to be determined.
Although this cost-effectiveness analysis is primarily performed for the Dutch situation, we believe the results are relevant for other situations as well. Although absolute participation rates and costs may differ among countries, the relative differences between CT colonography and colonoscopy primarily determine the comparative effectiveness. We suspect these relative differences in participation and costs differ less among countries. In a recent Italian study on comparative participation rates of CT colonography and colonoscopy (44), the difference between the two tests was similar to that in the COCOS trial (15% for colonoscopy vs 25% and 28% for CT colonography depending on the level of cathartic preparation). Furthermore, costs of CT colonography and colonoscopy make up a large part of personnel costs. Because it is likely that there is a constant ratio of CT colonography to colonoscopy personnel costs, the ratio between unit costs for CT colonography and colonoscopy will not be affected. Indeed, previous cost analyses found a similar ratio between CT colonography and colonoscopy Medicare reimbursement rates (34). Conversely, a study by Pyenson et al (45) found relatively lower costs for CT colonography, probably making CT colonography cost-effective also with equal participation rates. Furthermore, the use of the Dutch willingness-to-pay threshold (€20 000 per QALY gained) does not imply that our findings cannot be generalized, because CT colonography showed an ICER of €3200 per QALY gained compared with colonoscopy and many countries have an even higher willingness-to-pay threshold than the Dutch threshold. Another issue concerning generalizability is that CT colonography is not yet being performed on a large scale. This could affect performance on a community level. Indeed, a retrospective analysis in the English Bowel Cancer Screening Program and a Dutch study in which six physicians and three radiographers completed a CT colonography training program both showed that performance of CT colonography was higher at centers and for radiologists with more experience (46,47). However, the Dutch study estimated that the number of CT colonographic examinations needed to achieve sufficient performance was only 164 (46). Currently, the target population for CRC screening in the Netherlands encompasses approximately 4.5 million individuals. Therefore, we conclude that experience to achieve an adequate performance of CT colonography in an organized screening program can be met in a short period of time.
The implications of our study depend on the way screening is introduced. Our study conclusions that CT colonography is preferred over colonoscopy is most applicable in a national screening program offering a single test modality (which is usually the case in so-called organized screening programs). A comparison with alternative screening tests (eg, FIT), however, should also be updated in future research. Other options are also in place, mostly in opportunistic screening settings. An example of such an option is to offer participants a choice between screening tests. Because both colonoscopy and CT colonography are cost-effective in participants when compared with no screening, either test might be offered. Another option is to subsequently offer different screening modalities to nonparticipants (48), starting with the most cost-effective screening test for participants. Colonoscopy could then be offered initially, while nonparticipants could be offered CT colonography (or FIT). Three randomized controlled CRC screening trials that included a study arm offering a choice between screening tests found a participation rate in the group with choice as high as that in the group without choice (49–51). A Dutch study offering FIT to nonparticipants of sigmoidoscopy showed that the overall attendance of sigmoidoscopy plus FIT was still lower than that for FIT alone, whereas a similar Italian study showed the same attendance between those options (51). In summary, on the basis of the 56% higher CT colonography participation observed in the COCOS trial, CT colonography screening for CRC is more cost-effective than colonoscopy screening. The implementation of CT colonography screening requires prior satisfactory resolution of the optimal approach to managing extracolonic findings.
Implication for Patient Care
■ When deciding on single-modality screening for colorectal cancer, CT colonography is more cost-effective than colonoscopy screening on the basis of participation as observed in the Colonoscopy or Colonography for Screening trial.
Author Contributions
Author contributions: Guarantors of integrity of entire study, M.P.v.M., I.L.V., M.v.B.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; agrees to ensure any questions related to the work are appropriately resolved, all authors; literature research, M.P.v.M.; clinical studies, E.D.; statistical analysis, M.P.v.M., I.L.V., S.L.G., M.v.B.; and manuscript editing, all authors
References
- 1. . Global cancer statistics. CA Cancer J Clin 2011;61(2):69–90. Crossref, Medline, Google Scholar
- 2. . Fecal occult blood screening for colorectal cancer in the general population: results of a controlled trial. Cancer 1986;58(2):397–403. Crossref, Medline, Google Scholar
- 3. . Cochrane systematic review of colorectal cancer screening using the fecal occult blood test (hemoccult): an update. Am J Gastroenterol 2008;103(6):1541–1549. Crossref, Medline, Google Scholar
- 4. . Randomised study of screening for colorectal cancer with faecal-occult-blood test. Lancet 1996;348(9040):1467–1471. Crossref, Medline, Google Scholar
- 5. . Colorectal cancer mortality: effectiveness of biennial screening for fecal occult blood. J Natl Cancer Inst 1999;91(5):434–437. Crossref, Medline, Google Scholar
- 6. . Reduced incidence and mortality from colorectal cancer with flexible-sigmoidoscopy screening: a meta-analysis. World J Gastroenterol 2014;20(48):18466–18476. Crossref, Medline, Google Scholar
- 7. . Procedural success and complications of large-scale screening colonoscopy. Gastrointest Endosc 2002;55(3):307–314. Crossref, Medline, Google Scholar
- 8. . CT colonography with minimal bowel preparation: evaluation of tagging quality, patient acceptance and diagnostic accuracy in two iodine-based preparation schemes. Eur Radiol 2010;20(2):367–376. Crossref, Medline, Google Scholar
- 9. . CT colonography versus colonoscopy for the detection of advanced neoplasia. N Engl J Med 2007;357(14):1403–1412. Crossref, Medline, Google Scholar
- 10. . Diagnostic value of CT-colonography as compared to colonoscopy in an asymptomatic screening population: a meta-analysis. Eur Radiol 2011;21(8):1747–1763. Crossref, Medline, Google Scholar
- 11. . Community-based screening by colonoscopy or computed tomographic colonography in asymptomatic average-risk subjects. Am J Gastroenterol 2004;99(6):1145–1151. Crossref, Medline, Google Scholar
- 12. . Participation and yield of colonoscopy versus non-cathartic CT colonography in population-based screening for colorectal cancer: a randomised controlled trial. Lancet Oncol 2012;13(1):55–64. Crossref, Medline, Google Scholar
- 13. . Cost-effectiveness of computed tomography colonography in colorectal cancer screening: a systematic review. Int J Technol Assess Health Care 2012;28(4):415–423. Crossref, Medline, Google Scholar
- 14. . Cost-effectiveness of CT colonography. Radiol Clin North Am 2013;51(1):89–97. Crossref, Medline, Google Scholar
- 15. . CT colonography and cost-effectiveness. Eur Radiol 2008;18(11):2485–2497. Crossref, Medline, Google Scholar
- 16. . Unit costs in population-based colorectal cancer screening using CT colonography performed in university hospitals in The Netherlands. Eur Radiol 2013;23(4):897–907. Crossref, Medline, Google Scholar
- 17. . Population-based colorectal cancer screening by colonoscopy or CT-colonography [thesis]. Rotterdam, the Netherlands: Erasmus University Rotterdam, 2013. Google Scholar
- 18. . A novel hypothesis on the sensitivity of the fecal occult blood test: Results of a joint analysis of 3 randomized controlled trials. Cancer. 2009;115(11):2410–2419. Crossref, Medline, Google Scholar
- 19. . The MISCAN-COLON simulation model for the evaluation of colorectal cancer screening. Comput Biomed Res. 1999;32(1):13–33. Crossref, Medline, Google Scholar
- 20. . Statline. http://statline.cbs.nl/StatWeb/default.aspx. Accessed March 14, 2012. Google Scholar
- 21. . NVMDL. http://www.mdl.nl/uploads/240/1308/Richtlijn_Coloscopie_Surveillance_definitief_2013.pdf. Accessed September 25, 2014. Google Scholar
- 22. . Polyp miss rate determined by tandem colonoscopy: a systematic review. Am J Gastroenterol 2006;101(2):343–350. Crossref, Medline, Google Scholar
- 23. . Colorectal cancer: CT colonography and colonoscopy for detection—systematic review and meta-analysis. Radiology 2011;259(2):393–405. Link, Google Scholar
- 24. . High compliance rates observed for follow up colonoscopy post polypectomy are achievable outside of clinical trials: efficacy of polypectomy is not reduced by low compliance for follow up. Colorectal Dis 2004;6(3):158–161. Crossref, Medline, Google Scholar
- 25. . Random comparison of repeated faecal immunochemical testing at different intervals for population-based colorectal cancer screening. Gut 2013;62(3):409–415. Crossref, Medline, Google Scholar
- 26. . Utility valuations for outcome states of colorectal cancer. Am J Gastroenterol 1999;94(6):1650–1657. Crossref, Medline, Google Scholar
- 27. . The incidence and cost of unexpected hospital use after scheduled outpatient endoscopy. Arch Intern Med 2010;170(19):1752–1757. Crossref, Medline, Google Scholar
- 28. . Law and regulations. http://www.nza.nl/regelgeving/tarieven/. Accessed March 14, 2012. Google Scholar
- 29. . Farmacotherapeutisch Kompas. Diemen, the Netherlands: College voor zorgverzekeringen, 2014. Google Scholar
- 30. . The impact of healthcare costs in the last year of life and in all life years gained on the cost-effectiveness of cancer screening. Br J Cancer 2009;100(8):1240–1244. Crossref, Medline, Google Scholar
- 31. . Cost of care for elderly cancer patients in the United States. J Natl Cancer Inst 2008;100(9):630–641. Crossref, Medline, Google Scholar
- 32. . Guidelines for pharmacoeconomic studies: recommendations from the panel on cost effectiveness in health and medicine—panel on cost effectiveness in health and medicine. Pharmacoeconomics 1997;11(2):159–168. Crossref, Medline, Google Scholar
- 33. . Cost-effectiveness of prevention: opportunities for public health policy in the Netherlands [in Dutch]. Ned Tijdschr Geneeskd 2008;152(23):1329–1334. Medline, Google Scholar
- 34. . Cost-effectiveness of computed tomographic colonography screening for colorectal cancer in the Medicare population. J Natl Cancer Inst 2010;102(16):1238–1252. Crossref, Medline, Google Scholar
- 35. . At what costs will screening with CT colonography be competitive? A cost-effectiveness approach. Int J Cancer 2009;124(5):1161–1168. Crossref, Medline, Google Scholar
- 36. . Lower risk of advanced neoplasia among patients with a previous negative result from a fecal test for colorectal cancer. Gastroenterology 2012;142(3):497–504. Crossref, Medline, Google Scholar
- 37. . Results from the first three rounds of the Scottish demonstration pilot of FOBT screening for colorectal cancer. Gut 2009;58(4):530–535. Crossref, Medline, Google Scholar
- 38. . The UK colorectal cancer screening pilot: results of the second round of screening in England. Br J Cancer 2007;97(12):1601–1605. Crossref, Medline, Google Scholar
- 39. . Radiation-related cancer risks from CT colonography screening: a risk-benefit analysis. AJR Am J Roentgenol 2011;196(4):816–823. Crossref, Medline, Google Scholar
- 40. . Screening for colorectal cancer: updated evidence report and systematic review for the U.S. Preventive Services Task Force. JAMA 2016;315(23):2576–2594. Crossref, Medline, Google Scholar
- 41. . Revisiting Wilson and Jungner in the genomic age: a review of screening criteria over the past 40 years. Bull World Health Organ 2008;86(4):317–319. Crossref, Medline, Google Scholar
- 42. . Once-only flexible sigmoidoscopy screening in prevention of colorectal cancer: a multicentre randomised controlled trial. Lancet 2010;375(9726):1624–1633. Crossref, Medline, Google Scholar
- 43. . CT-colonography vs. colonoscopy for detection of high-risk sessile serrated polyps. Am J Gastroenterol 2016;111(4):516–522. Crossref, Medline, Google Scholar
- 44. . Reduced and full-preparation CT colonography, fecal immunochemical test, and colonoscopy for population screening of colorectal cancer: a randomized trial. J Natl Cancer Inst 2015;108(2):djv319. Crossref, Medline, Google Scholar
- 45. . Medicare cost of colorectal cancer screening: CT colonography vs. optical colonoscopy. Abdom Imaging 2015;40(8):2966–2976. Crossref, Medline, Google Scholar
- 46. . Evaluation of a standardized CT colonography training program for novice readers. Radiology 2011;258(2):477–487. Link, Google Scholar
- 47. . Use of CT colonography in the English Bowel Cancer Screening Programme. Gut 2014;63(6):964–973. Crossref, Medline, Google Scholar
- 48. . Uptake of faecal immunochemical test screening among nonparticipants in a flexible sigmoidoscopy screening programme. Int J Cancer 2012;130(9):2096–2102. Crossref, Medline, Google Scholar
- 49. . Adherence to colorectal cancer screening: a randomized clinical trial of competing strategies. Arch Intern Med 2012;172(7):575–582. Crossref, Medline, Google Scholar
- 50. . A comparison of colorectal neoplasia screening tests: a multicentre community-based study of the impact of consumer choice. Med J Aust 2006;184(11):546–550. Crossref, Medline, Google Scholar
- 51. . Randomized trial of different screening strategies for colorectal cancer: patient response and detection rates. J Natl Cancer Inst 2005;97(5):347–357. Crossref, Medline, Google Scholar
Article History
Received December 1, 2016; revision requested February 2, 2017; revision received September 11; accepted September 28; final version accepted November 14.Published online: Feb 27 2018
Published in print: June 2018
