With more than 1.2 million new diagnoses and more than 600 000 deaths per year, colorectal cancer (CRC) is the third most common cancer worldwide and the fourth most common cause of cancer-related death in women and men (Ferlay et al., 2013). Results from randomized trials on the effectiveness of endoscopic screening are available for sigmoidoscopy only, but evidence from observational studies suggests that the majority of CRC cases and deaths could be prevented by screening colonoscopy with detection and removal of colorectal adenomas (Brenner et al., 2014a). Colonoscopy is recommended for CRC screening by expert panels in a number of countries, including the USA and Germany (Rex et al., 2009; Qaseem et al., 2012; European Colorectal Cancer Screening Guidelines Working Group et al., 2013; Pox et al., 2013).
High quality of screening colonoscopy is mandatory to achieve the maximum possible CRC prevention. This requires not only appropriate technical equipment but also experience of physicians. The adenoma detection rate has been suggested as an indicator of the quality of screening colonoscopy (Kaminski et al., 2010). Reported adenoma detection rates have varied widely across populations (Barclay et al., 2006; Church, 2008; Millan et al., 2008; Benson et al., 2010; Kaminski et al., 2010; Ferlitsch et al., 2011; Adler et al., 2012; Coriat et al., 2012; Pox et al., 2012; Barret et al., 2013; Greenspan et al., 2013; Corley et al., 2014), but this variation might partly reflect differences in neoplasm prevalences between those populations. A few studies specifically focused on interphysician variation in detection rates within the same study population, showing a wide range of adenoma detection rates among colonoscopists (Millan et al., 2008; Adler et al., 2012; Greenspan et al., 2013; Corley et al., 2014). However, two of those studies were based on small groups of physicians (six and 19, respectively) from single US medical centers (Millan et al., 2008; Greenspan et al., 2013). The most recent study included 136 gastroenterologists serving patients of a specific healthcare delivery organization (Corley et al., 2014). The study reported an inverse association between adenoma detection rate and the occurrence of interval cancers, which supports the use of this measure as a quality indicator of screening colonoscopy.
The aim of this study was to investigate interphysician variation in neoplasm detection rates in routine practice in the German healthcare system and, more specifically, to assess possible differences in the detection rates of any neoplasms (including nonadvanced adenoma, advanced adenoma, and CRC), in advanced neoplasm detection rates (including advanced adenoma and CRC), as well as in CRC detection rates according to volume of screening colonoscopies. Disclosure of such differences would be of high relevance for the definition of quality criteria and the requirements for quality assurance of screening colonoscopy.
Materials and methods
Colonoscopy has been offered as a primary screening investigation for CRC in the German healthcare system since October 2002 (Pox et al., 2012). Men and women are entitled to have a first screening colonoscopy from age 55 on. If the first screening colonoscopy was conducted before age 65, another screening colonoscopy is offered 10 or more years after the first screening colonoscopy. For this analysis, data of the quality assurance program of the Bavarian Association of Statutory Health Insurance Physicians (‘Qualitätsmaßnahme Koloskopie’) were used, which was run from 2006 to 2009. As documentation of the quality assurance program partially changed in 2006, including changes in the specification of histological findings and definition of adenoma size, this study includes data collected between 2007 and 2009 only. Bavaria is a federal state located in the south of Germany (∼12.5 million inhabitants, 15% of the German population). Between 2007 and 2009, 89% of the Bavarian population was insured by the Statutory Health Insurance system (German Federal Ministry of Health, 2013; Health insurance companies [Krankenkassen Deutschland], 2013)). Offers of screening and nonscreening colonoscopies in Bavaria are the same as in other German states; however, comprehensive monitoring of both screening and nonscreening colonoscopies in a state-wide registry was done in Bavaria only.
Quality assurance of colonoscopy in Germany covers the professional and technical requirements for the execution of screening or other (nonscreening) colonoscopies. Only certified specialist physicians in internal medicine with specialization in gastroenterology, as well as surgeons, are entitled to perform screening colonoscopies. To maintain certification, a minimum of 200 total colonoscopies, including 10 polypectomies per year, are required.
The database, which has been described in detail previously (Crispin et al., 2009; Kolligs et al., 2011; Crispin et al., 2013; Stock et al., 2013a, 2013b; Brenner et al., 2014b, 2014c), contains electronic documentation of screening and nonscreening colonoscopies between 2006 and 2009, including information on demographics, process quality, findings, complications, and diagnoses and treatments. For privacy reasons, however, no further characteristics of endoscopists (such as age, sex, specialty, setting, etc.) or information on the use of (new) screening technologies were available.
Inclusion and exclusion criteria
Individuals aged at least 55 years, who underwent screening colonoscopy between 2007 and 2009 in Bavaria, Germany, were included in the present analysis. The lower age limit is defined by the offer of screening colonoscopy in Germany. Specification of histological findings and definition of adenoma size, needed to categorize findings at colonoscopy, were not available in 2006. Thus, individuals who underwent screening colonoscopy in that year were excluded from this study.
To compare interphysician variation in diagnostic colonoscopies, the same physicians who performed screening colonoscopies between 2007 and 2009 were selected. There was no age limit for individuals who underwent diagnostic colonoscopy.
Findings at screening colonoscopy were categorized into three groups. The first group comprised ‘any neoplasm’, including nonadvanced adenomas (1–3 tubular adenomas, each <1 cm, only low-grade intraepithelial neoplasia), advanced adenomas (≥4 tubular adenomas, ≥1 adenoma ≥1 cm, ≥1 adenoma with tubulo-villous or villous structure, high-grade intraepithelial neoplasia), and CRC. The second group comprised ‘advanced neoplasm’, including advanced adenomas and CRC, and the third group comprised CRC only. Note that usually the prevalence of 3 or more tubular adenomas is considered a high-risk adenoma situation. However, the reporting forms only differentiated between individuals with 0, 1–3, and at least 4 adenomas. Participants were classified according to the most advanced finding at colonoscopy.
Descriptive data on the study population are presented in absolute and relative frequencies. Percentages and corresponding 95% confidence intervals (CI) of colonoscopies with detection of any neoplasm, any advanced neoplasm, and CRC were determined per physician. Physicians were grouped according to the number of screening colonoscopies performed per year (<50, 50–99, 100–199, ≥200 colonoscopies). Categories were chosen to include approximately one-fourth of physicians each. These numbers refer to screening colonoscopies only; the overall number of colonoscopies conducted by the endoscopists was more than three-fold higher. Alternative categorization by exact quartiles yielded very similar results.
Spearman’s rank correlation coefficients were calculated to quantify the correlation between screening colonoscopy volume and detection rate of any neoplasm among physicians. Log-binomial regression was performed to estimate relative detection rates (RDRs) (Spiegelman and Hertzmark, 2005) and to test for trends in the detection of any neoplasm, any advanced neoplasm, and CRC across the four groups of physicians, adjusting for sex and age (as a continuous variable), and accounting for clustering according to physicians.
Sex-specific analyses were carried out in addition to analyses on screening colonoscopy volume for the entire study population.
Using the same physician groups defined by screening colonoscopy volume, we also analyzed detection rates during diagnostic colonoscopies to compare interphysician variation. Diagnostic colonoscopies were performed for the following indications: macroscopically visible blood in stool, positive occult blood test, surveillance after preceding adenoma or CRC diagnosis, and clinical symptoms such as diarrhea, abdominal pain, or constipation.
Statistical significance was defined by P less than 0.05 (two-sided tests). Analyses were performed using the statistics software SAS©, version 9.2 (SAS Institute Inc., Cary, North Carolina, USA). PROC GENMOD was used for log-binomial regression models. The REPEATED SUBJECT-statement was used to account for clustering by physicians. To assess dose–response relationships between screening colonoscopy volume and adenoma detection rates cubic spline regression models were fitted using the ‘mgcv’-package (Wood, 2015) in R language and environment for statistical computing, version 3.1.2 (R Core Team, 2015).
The ethics committee of the Medical Faculty at the University of Heidelberg has approved this study.
In total, 203 363 individuals aged at least 55 years (median age 64 years, 55.2% women) who underwent screening colonoscopy between 2007 and 2009 in Bavaria, Germany, were included in the database. Screening colonoscopies were performed by 509 physicians. Inadequate bowel cleaning and incompleteness of colonoscopies were reported for 0.3 and 1.5% of colonoscopies, respectively.
Overall, 54 801 individuals (26.9%) were found to have any neoplasm, of whom 19 814 (9.7%) had at least one advanced neoplasm (advanced adenoma or CRC). CRC was detected in 2318 individuals (1.1%).
In total, 132 physicians performed fewer than 50 screening colonoscopies per year, 106 physicians performed 50–99 screening colonoscopies per year, 146 physicians performed 100–199 screening colonoscopies per year, and 125 physicians performed 200 or more screening colonoscopies per year (Table 1). However, only ∼4% of screening colonoscopies (n=8107) were conducted by the first group (<50 screening colonoscopies per year), whereas more than half of the screening colonoscopies were conducted by the fourth group (≥200 screening colonoscopies per year). Detection rates of any neoplasm increased with volume of screening colonoscopies from 21.86% (95% CI 20.96–22.76%) in the group of physicians with fewer than 50 screening colonoscopies per year to 27.51% (95% CI 27.25–27.76%) among physicians conducting at least 200 screening colonoscopies per year (P for trend <0.0001). In particular, detection of any neoplasm was 17% lower (RDR 0.83, 95% CI 0.76–0.90) in the group of physicians conducting fewer than 50 screening colonoscopies per year compared with physicians who performed more than 200 screening colonoscopies per year. Increases in detection rates with colonoscopy volume, with clearly lower detection rates for physicians performing fewer than 50 screening colonoscopies per year, were also seen for advanced neoplasms and for CRC (P-values for trend 0.16 and <0.0001, respectively). Detection rates did not increase further with screening colonoscopy volume among physicians who performed more than 200 screening colonoscopies per year.
In sex-specific analyses, detection rates were consistently higher in men than in women (Table 2). The detection rates of any neoplasm increased with the volume of screening colonoscopies from 27.50% (95% CI 26.11–28.89%) in men and 16.56% (95% CI 15.43–17.70%) in women for the group of physicians with fewer than 50 screening colonoscopies per year to 34.83% (95% CI 34.42–35.24%) in men and 21.78% (95% CI 21.47–22.10%) in women among physicians conducting at least 200 screening colonoscopies per year (P for trend each <0.0001). Trends of increasing detection rates with increasing colonoscopy volume were also seen when the analysis was restricted to detection of advanced neoplasms or CRC, even though these trends were only statistically significant for CRC in sex-specific analyses (men: P for trend 0.0003; women: P for trend 0.019). Compared with physicians conducting at least 200 screening colonoscopies per year, the detection rate of any neoplasm was 23% lower (RDR 0.77, 95% CI 0.69–0.87) in men and 24% lower (RDR 0.76, 95% CI 0.66–0.87) in women in the group of physicians conducting fewer than 50 screening colonoscopies per year.
Neoplasm detection rates according to screening colonoscopy volume by individual physicians are shown in Fig. 1. Although a positive correlation between colonoscopy volume and neoplasm detection rate is evident (Spearman’s rank correlation coefficient 0.25, P<0.0001), the figure also shows that there is very large variation in neoplasm detection rates among physicians with comparable colonoscopy volume. The same patterns were observed for advanced neoplasm detection rates (Spearman’s rank correlation coefficient 0.25, P<0.0001) according to screening colonoscopy volume by individual physicians (Supplementary Fig. 1, Supplemental digital content 1, http://links.lww.com/EJCP/A41).
Using the same physician groups defined by screening colonoscopy volume, we also analyzed detection rates during diagnostic colonoscopies. The detection rates of any neoplasm in diagnostic colonoscopies increased with the volume of screening colonoscopies from 19.62% (95% CI 19.32–19.91%) in the group of physicians with fewer than 50 screening colonoscopies per year to 21.16% (95% CI 21.02–21.30%) among physicians conducting at least 200 screening colonoscopies per year (P for trend <0.0001) (Supplementary Table 1, Supplemental digital content 2, http://links.lww.com/EJCP/A42). There was no evidence that detection rates of advanced neoplasms and CRC in diagnostic colonoscopies were related to colonoscopy volume (P-values for trend 0.62 and 0.37, respectively).
In this registry-based study, the association of screening colonoscopy volume with colorectal neoplasm detection rate was investigated. The analysis was based on a large, state-wide documentation of screening colonoscopies from Bavaria, Germany, and reflects routine practice in Germany in the time period from 2007 to 2009. We observed increasing detection rates of any neoplasm (including nonadvanced adenoma, advanced adenoma, and CRC) with increasing screening colonoscopy volume. Trends of increasing detection rates with increasing screening colonoscopy volume were also seen for advanced neoplasm and CRC. Overall, detection rates were higher for male patients than for female patients, but patterns of increasing detection rates with increasing screening colonoscopy volume were very similar for both sexes.
Between-group differences in neoplasm detection rates were highest between physicians conducting fewer than 50 screening colonoscopies per year and physicians with higher colonoscopy volumes. Among physicians conducting at least 50 screening colonoscopies per year, further increases in detection rates were rather marginal. These patterns suggest that 50 screening colonoscopies per year might be a reasonable minimum number for high-quality screening colonoscopy. Although the number of procedures performed annually is not a reliable measure of quality, achieving an adequate volume is essential for maintaining skills and effectively monitoring performance. It should be noted, however, that colonoscopy volume in our study considered colonoscopies conducted as primary screening colonoscopy examinations only. In Germany, the total number of colonoscopies conducted in outpatient settings is approximately four-fold higher than the number of screening colonoscopies. Assuming similar proportions of screening colonoscopies across the groups of physicians compared, the critical total number of colonoscopies conducted per year might therefore be ∼200, which is equivalent to the minimum required number of colonoscopies per year to maintain certification.
A wide range of adenoma detection rates among screening colonoscopists has been reported in other studies (Millan et al., 2008; Adler et al., 2012; Chokshi et al., 2012; Greenspan et al., 2013; Menees et al., 2013; Corley et al., 2014). Besides colonoscopy volume, other individual colonoscopist factors, such as physicians’ experience, setting (hospital or private practice), or withdrawal time, as well as bowel preparation quality, sedation, and instrument quality, seem to play a major role in the quality of screening colonoscopy. This is supported by the strong variation in adenoma detection rates as seen at all levels of screening colonoscopy volumes in our study (Fig. 1). As variables in this state-wide database were limited, reasons for this variation can only be speculated about. There might be differences in physicians’ experience, especially in regard to their professional training. In addition, there might be some cases in which a higher number of performed colonoscopies only stands for quicker examination time and less attentiveness. The importance of the adenoma detection rate as a quality benchmark of colonoscopy is highlighted by the recent findings of Corley et al. (2014), who demonstrated a clear inverse association between adenoma detection rate and the occurrence of interval cancers diagnosed 6 months to 10 years after screening colonoscopy.
Although the overall colorectal neoplasm detection rates in our study are in agreement with the recommended limit of at least 20% reported in the literature (Rex et al., 2006; Kaminski et al., 2010), true prevalences of colorectal neoplasms are likely to be higher because of non-negligible miss rates, which are likely to account for most of the differences seen by colonoscopy volume. In a systematic review published in 2006, van Rijn et al. reported miss rates of 21% for very small adenomas (≤5 mm), 13% for small adenomas (6–9 mm), and 2% for large adenomas (≥10 mm). Substantially larger miss rates have been suggested by more recent studies (Chokshi et al., 2012; Leufkens et al., 2012; Menees et al., 2013).
Potential causes for colonoscopy failure to detect neoplasms and to prevent CRC include, among others, inadequate bowel preparation, failure to reach the cecum, abbreviated withdrawal of colonoscopy, and failure to detect sessile or flat lesions (Harewood et al., 2003; Froehlich et al., 2005; Aslinia et al., 2006; Barclay et al., 2006; Rex et al., 2006; Chen and Rex, 2007; Heresbach et al., 2008; Soetikno et al., 2008; Ibáñez et al., 2011; Lebwohl et al., 2011; Chokshi et al., 2012). In the present study, inadequate bowel cleaning was rare (0.3%), and the cecum was reached in 98.5% of all individuals. This high completion rate is well above the benchmarks set by expert panels, such as those defined in the European guidelines for quality and diagnosis (European Colorectal Cancer Screening Guidelines Working Group et al., 2013), the UK National Health Service Bowel Cancer Screening Programme (Rutter and Chilton, 2011), or the recommendations of the US Multi-Society Task Force on Colorectal Cancer (Rex et al., 2009). The major differences in neoplasm detection rates according to colonoscopy volume suggest that non-negligible proportions of adenomas are likely to be missed in routine practice even with such high completion rates.
Our study has specific strengths and limitations. Strengths include the large colonoscopy documentation database with almost complete coverage of a population of 12.5 million people. With 509 physicians and 203 363 individuals who underwent screening colonoscopy, this study represents to our knowledge the largest study assessing the association of screening colonoscopy volume with colorectal neoplasm detection rate reported to date. All participating physicians demonstrated a very good bowel preparation quality of their patients and a high completion rate that was well above the benchmarks set by expert panels.
Limitations of this study mostly result from the limited number of variables included in this state-wide database. First, no information about physicians’ experience, specialization, and setting (hospital or private practice) was available, which might be important factors in the detection of CRC and its precancerous lesions (Adler et al., 2012). Second, examination or withdrawal times of screening colonoscopies were not known, which have been shown to be associated with neoplasm detection rates (Barclay et al., 2006; Simmons et al., 2006; Barclay et al., 2008; Benson et al., 2010; Lee et al., 2013). Therefore, it was not possible to assess the impact of these specific determinants of the colorectal neoplasm detection rate identified in previous studies. Third, information on patient risk factors, such as smoking, BMI, or diabetes (Reid et al., 2003; Jacobs et al., 2007; Sedjo et al., 2007), which are known to be associated with the risk of colorectal neoplasms and which have also been associated with neoplasm detection rates in some studies, was not available. Finally, the potential for misclassification of colorectal lesions by the physician has to be kept in mind.
Despite these limitations, our results support a potential role of a minimum colonoscopy volume for colonoscopists for high-quality colonoscopy screening. Differences in neoplasm detection rates were high between physicians conducting fewer than 50 screening colonoscopies per year and physicians with higher screening colonoscopy volume. These results support the definition of a minimum colonoscopy volume, besides other quality criteria, for certification for screening colonoscopy. Further research should take into account the influence of case volume and other parameters, such as physicians’ specialty (gastroenterologists or surgeons), their age, and previous experience, for defining the most effective measures to ensure maximum possible quality standards for screening colonoscopy and to minimize occurrence of interval cancers.
The authors are grateful to the Bavarian Association of Statutory Health Insurance Physicians for providing the colonoscopy documentation data used in the present study.
This study was supported in part by the German Research Council (Deutsche Forschungsgemeinschaft), project no. BR 1704/6-6.
Author contributions: Conception and design: Hermann Brenner; acquisition of data: Berndt Birkner, Christian Stock, Hermann Brenner; analysis and interpretation of the data: all authors; drafting of the article: Nadine Zwink; critical revision of the article for important intellectual content: all authors; statistical analysis: Nadine Zwink; final approval of the article: all authors.
Conflicts of interest
There are no conflicts of interest.
Adler A, Wegscheider K, Lieberman D, Aminalai A, Aschenbeck J, Drossel R, et al (2012). Factors determining the quality of screening colonoscopy: a prospective study on adenoma detection rates, from 12 134 examinations (Berlin colonoscopy project 3, BECOP-3). Gut 62:236–241.
Aslinia F, Uradomo L, Steele A, Greenwald BD, Raufman JP (2006). Quality assessment of colonoscopic cecal intubation: an analysis of 6 years of continuous practice at a university hospital. Am J Gastroenterol 101:721–731.
Barclay RL, Vicari JJ, Doughty AS, Johanson JF, Greenlaw RL (2006). Colonoscopic withdrawal times and adenoma detection during screening colonoscopy. N Engl J Med 355:2533–2541.
Barclay RL, Vicari JJ, Greenlaw RL (2008). Effect of a time-dependent colonoscopic withdrawal protocol on adenoma detection during screening colonoscopy. Clin Gastroenterol Hepatol 6:1091–1098.
Barret M, Boustiere C, Canard JM, Arpurt JP, Bernardini D, Bulois P, et al, Société Française d’Endoscopie Digestive (2013). Factors associated with adenoma detection rate and diagnosis of polyps and colorectal cancer during colonoscopy in France: results of a prospective, nationwide survey. PLoS One 8:e68947.
Benson ME, Reichelderfer M, Said A, Gaumnitz EA, Pfau PR (2010). Variation in colonoscopic technique and adenoma detection rates at an academic gastroenterology unit. Dig Dis Sci 55:166–171.
Brenner H, Stock C, Hoffmeister M (2014a). Effect of screening sigmoidoscopy and screening colonoscopy on colorectal cancer incidence and mortality: systematic review and meta-analysis of randomised controlled trials and observational studies. BMJ 348:g2467.
Brenner H, Hoffmeister M, Birkner B, Stock C (2014b). Men with negative results of guaiac-based fecal occult blood test have higher prevalences of colorectal neoplasms than women with positive results. Int J Cancer 134:2927–2934.
Brenner H, Hoffmeister M, Birkner B, Stock C (2014c). Diagnostic performance of guaiac-based fecal occult blood test in routine screening: state-wide analysis from Bavaria, Germany. Am J Gastroenterol 109:427–435.
Chen SC, Rex DK (2007). Endoscopist can be more powerful than age and male gender in predicting adenoma detection at colonoscopy. Am J Gastroenterol 102:856–861.
Chokshi RV, Hovis CE, Hollander T, Early DS, Wang JS (2012). Prevalence of missed adenomas in patients with inadequate bowel preparation on screening colonoscopy. Gastrointest Endosc 75:1197–1203.
Church J (2008). Adenoma detection rate and the quality of colonoscopy: the sword has two edges. Dis Colon Rectum 51:520–523.
Coriat R, Lecler A, Lamarque D, Deyra J, Roche H, Nizou C, et al (2012). Quality indicators for colonoscopy procedures: a prospective multicentre method for endoscopy units. PLoS One 7:e33957.
Corley DA, Jensen CD, Marks AR, Zhao WK, Lee JK, Doubeni CA, et al (2014). Adenoma detection rate and risk of colorectal cancer and death. N Engl J Med 370:1298–1306.
Crispin A, Birkner B, Munte A, Nusko G, Mansmann U (2009). Process quality and incidence of acute complications in a series of more than 230 000 outpatient colonoscopies. Endoscopy 41:1018–1025.
Crispin A, Mansmann U, Munte A, Op den Winkel M, Göke B, Kolligs FT (2013). A direct comparison of the prevalence of advanced adenoma and cancer between surveillance and screening colonoscopies. Digestion 87:170–175.
Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, et al (2013). GLOBOCAN 2012 v10, cancer incidence and mortality worldwide: IARC CancerBase No 11. Lyon, France: International Agency for Research on Cancer.
Ferlitsch M, Reinhart K, Pramhas S, Wiener C, Gal O, Bannert C, et al (2011). Sex-specific prevalence of adenomas, advanced adenomas, and colorectal cancer in individuals undergoing screening colonoscopy. JAMA 306:1352–1358.
Froehlich F, Wietlisbach V, Gonvers JJ, Burnand B, Vader JP (2005). Impact of colonic cleansing on quality and diagnostic yield of colonoscopy: the European Panel of Appropriateness of Gastrointestinal Endoscopy European multicenter study. Gastrointest Endosc 61:378–384.
German Federal Ministry of Health (2013). Statutory health insurance (SHI). Available at: http://www.bmg.bund.de
. [Accessed 10 September 2013].
Greenspan M, Rajan KB, Baig A, Beck T, Mobarhan S, Melson J (2013). Advanced adenoma detection rate is independent of nonadvanced adenoma detection rate. Am J Gastroenterol 108:1286–1292.
Harewood GC, Sharma VK, de Garmo P (2003). Impact of colonoscopy preparation quality on detection of suspected colonic neoplasia. Gastrointest Endosc 58:76–79.
Heresbach D, Barrioz T, Lapalus MG, Coumaros D, Bauret P, Potier P, et al (2008). Miss rate for colorectal neoplastic polyps: a prospective multicenter study of back-to-back video colonoscopies. Endoscopy 40:284–290.
Ibáñez M, Parra-Blanco A, Zaballa P, Jiménez A, Fernández-Velázquez R, Fernández-Sordo JO, et al (2011). Usefulness of an intensive bowel cleansing strategy for repeat colonoscopy after preparation failure. Dis Colon Rectum 54:1578–1584.
Jacobs ET, Martínez ME, Alberts DS, Jiang R, Lance P, Lowe KA, Thompson PA (2007). Association between body size and colorectal adenoma recurrence. Clin Gastroenterol Hepatol 5:982–990.
Kaminski MF, Regula J, Kraszewska E, Polkowski M, Wojciechowska U, Didkowska J, et al (2010). Quality indicators for colonoscopy and the risk of interval cancer. N Engl J Med 362:1795–1803.
Kolligs FT, Crispin A, Munte A, Wagner A, Mansmann U, Göke B (2011). Risk of advanced colorectal neoplasia according to age and gender. PLoS One 6:e20076.
Lebwohl B, Kastrinos F, Glick M, Rosenbaum AJ, Wang T, Neugut AI (2011). The impact of suboptimal bowel preparation on adenoma miss rates and the factors associated with early repeat colonoscopy. Gastrointest Endosc 73:1207–1214.
Lee TJ, Blanks RG, Rees CJ, Wright KC, Nickerson C, Moss SM, et al (2013). Longer mean colonoscopy withdrawal time is associated with increased adenoma detection: evidence from the Bowel Cancer Screening Programme in England. Endoscopy 45:20–26.
Leufkens AM, van Oijen MG, Vleggaar FP, Siersema PD (2012). Factors influencing the miss rate of polyps in a back-to-back colonoscopy study. Endoscopy 44:470–475.
Menees SB, Kim HM, Elliott EE, Mickevicius JL, Graustein BB, Schoenfeld PS (2013). The impact of fair colonoscopy preparation on colonoscopy use and adenoma miss rates in patients undergoing outpatient colonoscopy. Gastrointest Endosc 78:510–516.
Millan MS, Gross P, Manilich E, Church JM (2008). Adenoma detection rate: the real indicator of quality in colonoscopy. Dis Colon Rectum 51:1217–1220.
Pox CP, Altenhofen L, Brenner H, Theilmeier A, Von Stillfried D, Schmiegel W (2012). Efficacy of a nationwide screening colonoscopy program for colorectal cancer. Gastroenterology 142:1460.e2–1467.e2.
Pox C, Aretz S, Bischoff SC, Graeven U, Hass M, Heußner P, et al, Leitlinienprogramm Onkologie der AWMF; Deutschen Krebsgesellschaft e. V; Deutschen Krebshilfe e. V (2013). S3-guideline colorectal cancer version 1.0. Z Gastroenterol 51:753–854.
Qaseem A, Denberg T, Hopkins RJ, Humphrey L, Levine J, Sweet D, et al (2012). Screening for colorectal cancer: a guidance statement from the American College of Physicians. Ann Intern Med 156:378–386.
R Core Team (2015). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available at: http://www.R-project.org/
. [Accessed 17 November 2015].
Reid ME, Marshall JR, Roe D, Lebowitz M, Alberts D, Battacharyya AK, Martinez ME (2003). Smoking exposure as a risk factor for prevalent and recurrent colorectal adenomas. Cancer Epidemiol Biomarkers Prev 12:1006–1011.
Rex DK, Petrini JL, Baron TH, Chak A, Cohen J, Deal SE, et al (2006). Quality indicators for colonoscopy. Am J Gastroenterol 101:873–885.
Rex DK, Johnson DA, Anderson JC, Schoenfeld PS, Burke CA, Inadomi JM, American College of Gastroenterology (2009). American College of Gastroenterology guidelines for colorectal cancer screening 2009 [corrected]. Am J Gastroenterol 104:739–750.
Sedjo RL, Byers T, Levin TR, Haffner SM, Saad MF, Tooze JA, D’Agostino RB Jr (2007). Change in body size and the risk of colorectal adenomas. Cancer Epidemiol Biomarkers Prev 16:526–531.
Simmons DT, Harewood GC, Baron TH, Petersen BT, Wang KK, Boyd-Enders F, Ott BJ (2006). Impact of endoscopist withdrawal speed on polyp yield: implications for optimal colonoscopy withdrawal time. Aliment Pharmacol Ther 24:965–971.
Soetikno RM, Kaltenbach T, Rouse RV, Park W, Maheshwari A, Sato T, et al (2008). Prevalence of nonpolypoid (flat and depressed) colorectal neoplasms in asymptomatic and symptomatic adults. JAMA 299:1027–1035.
Spiegelman D, Hertzmark E (2005). Easy SAS calculations for risk or prevalence ratios and differences. Am J Epidemiol 162:199–200.
Stock C, Hoffmeister M, Birkner B, Brenner H (2013a). Performance of additional colonoscopies and yield of neoplasms within 3 years after screening colonoscopy: a historical cohort study. Endoscopy 45:537–546.
Stock C, Hoffmeister M, Birkner B, Brenner H (2013b). Inter-physician variation in follow-up colonoscopies after screening colonoscopy. PLoS One 8:e69312.
van Rijn JC, Reitsma JB, Stoker J, Bossuyt PM, van Deventer SJ, Dekker E (2006). Polyp miss rate determined by tandem colonoscopy: a systematic review. Am J Gastroenterol 101:343–350.
von Karsa L, Patnick J, Segnan N, Atkin W, Halloran S, Lansdorp-Vogelaar I, et al, European Colorectal Cancer Screening Guidelines Working Group (2013). European guidelines for quality assurance in colorectal cancer screening and diagnosis: overview and introduction to the full supplement publication. Endoscopy 45:51–59.