Population Screening for colorectal cancer (DTB 2006)
Colorectal cancer is the second commonest cause of cancer death in the UK ,accounting for around 16,000 deaths per year.(Cancer research UK 2006) Most patients with colorectal cancer are over the age of 60 years at diagnosis; however, almost 3,000 patients (14.7%) of those diagnosed in England in 2003 were aged under 60 years. (Office for National Statistics 2005). In England and Wales, around 55% of patients who have colorectal cancer present with advanced disease that has either spread to lymph nodes, metastasised to other organs or is so locally advanced where surgery is unlikely to be curative. (DH 2003) Overall 5-year survival is said to vary from, for example, 45-50% in the UK to 64.1% in the USA (Coleman 1999, SEER 2006) but of those who present with metastatic colorectal cancer 5-year survival is 7% at most in the UK, despite the use of combination chemotherapy. (DTB May 2006)
A national bowel screening program is being introduced in England and Scotland. Here we assess the role of screening in reducing mortality from colorectal cancer in a general population.
What are the criteria for population screening?
Most colorectal cancers are thought to arise from adenomatous polyps that develop over a period of years. An ideal screening programme for colorectal cancer would be able to detect either early stage curable cancers or detect and remove adenomatous polyps. Any test used for population screening needs to be sensitive enough to detect this early disease if it is present (low false negatives); specific enough to ensure that people do not have unnecessary extra investigations (low false positives); and be cheap, reproducible, and safe enough to be acceptable on a population basis. (Wagner 1996) Currently, the most effective tests for bowel cancer screen mainly for early cancers that can be more easily treated. The two main methods used include stool testing to detect early signs of bleeding from cancers or adenomas, or endoscopy. Although none of these modalities has all the requirements for an ideal test, there is a substantial amount of data about their effectiveness in screened populations.
Randomised controlled trials are essential before implementing a population-wide screening programme due to a number of potential biases that can appear to improve survival from cancer even when the screening is not effective. (Sackett 1991) ‘Lead time bias’ occurs when a cancer is detected earlier due to screening but may not be any more treatable. However, the cancer will automatically appear to have a longer observed course because it was detected earlier. ‘Patient selection bias’ occurs when people who volunteer for screening do not accurately represent the population to be screened. They are often healthier than an average population. (Simon 1984) ‘Length bias’ occurs initially with screening and is where at one point in time a disproportionate number of slower growing (and more curable) malignancies will be detected, rather than those with shorter more aggressive histories. (Feinleib 1969) For these reasons, it is important in trials of screening to use absolute reduction in death from cancer rather than mean survival as the outcome measure.
Adenoma to adenocarcinoma?
Adenomas can be described as tubular (most common); tubulovillous; or villous adenomas (least common).(Konishi 1982) In addition, all adenomas can be graded by histological and cytological criteria into three grades of epithelial dysplasia; mild, moderate and severe. It is common to see differing grades of dysplasia in any one adenoma and severe dysplasia is associated with a much higher risk of malignancy. (Muto 1975) As the histology of adenomas become more villous, or they become larger or more frequent, the grade of dysplasia increases, with size the most practical indicator of dysplasia. (Konishi 1982) One study showed a risk of malignancy of only 1 in 100 with adenomas under 1cm in size; of 1 in 10 over 2cm size; and of over 50% in those over 2cm in size. (Muto 1975)
The evidence of the benefit of removing colorectal polyps is already established in those people with diseases at high-risk of colorectal cancer such as familial adenomatous polyposis (FAP) (Belchetz 1996, Bulow 1995) and hereditary non-polyposis colorectal cancer (NPCC), (Morton 1993, Jarvinen 2000, Vasen 1995) There is also some evidence that removing polyps in patients in the general population is effective. (Mandel 2000) Screening for high risk populations, including those with inflammatory bowel disease or a strong family history of bowel cancer, is not discussed here.
Faecal occult blood testing (FOBT)
Faecal occult blood tests are the mostly widely used screening test for bowel cancer, leading to reductions in mortality by aiding detection of early, more treatable, cancers. A phenolic compound in the blood undergoes oxidation to a quinone structure that changes colour. (Simon 1984) In the Hemoccult test, for example, a test paper is impregnated with gum guaiac with hydrogen peroxide as a developer solution. Sensitivity can be increased by rehydration, where distilled water is added to the slides just before the reagents are added; however specificity is reduced.(Pignone 2002) The sensitivity of the test is reduced by a number of factors; the peroxidase-like activity of haemoglobin can be reduced as it passes through the gut; the normal stool can contain blood; (Doran 1982, Herzog 1982) and cancers do not bleed continuously; all leading to increased numbers of false negatives. (Doran 1982) The result is therefore dependent upon the number of times the test is repeated (Farrands 1983, Pye 1986) and the most common protocol for guiuac-based tests requires 2 separate samples from 3 consecutive stools (a total of six samples). Some trials have recommended dietary restrictions around the time of testing as certain foods have been thought to increase the number of false positives with unrehydrated Hemoccult tests. However a meta-analysis in 2001 found that food restriction did not appear to change positivity rates. (Pignone 2001)
Efficacy of faecal occult-blood testing
A systematic review and meta-analysis, in 1998, (Towler 1998) looked at the results of the six major controlled trials of screening for colorectal cancer using the Hemoccult faecal occult blood test (mostly for annual or biennial screening) involving 442,000 people aged 40 years or over. The trials in Minnesota, (Mandel 1993) Nottingham, (Hardcastle 1996) Funen, (Kronborg 1996) and Gothenburg, (Kewenter 1994) were all randomised. The Burgundy, (Faivre 2004) and New York trials (Winawer 1993) were subject to non-randomised allocation. In addition, the New York trial offered sigmoidoscopy to all participants. Patients with a positive Hemoccult test were referred for further investigation, mainly colonoscopy or sigmoidoscopy with double contrast barium enema. In most trials, between 2-6% of the screened group had at least one endoscopy (colonoscopy or sigmoidoscopy). In the Minnesota trial, 28-38% of the trial participants had colonoscopy.
In the meta-analysis, the sensitivity of the Hemoccult test varied from 46% (unhydrated) to 92% (rehydrated), where sensitivity was the proportion of all colorectal cancers which were detected by screening and where all colorectal cancers was the sum of screening detected cancers (true positives) and interval cancers within one or two years of screening (false negatives). (Towler 1998) Between 60 and 90% of all those in the screening groups completed at least one Hemoccult test. The meta-analysis showed that screening resulted in a significant overall reduction in mortality from colorectal cancer of 16% (relative risk 0.84 [0.77 to 0.93]).When the relative risk was adjusted for attendance for screening the overall relative reduction in mortality was 23% (RR 0.77 [0.57 to 0.89]). The number needed to screen in order to prevent one death from colorectal cancer over 10 years was 1173 (741 to 2807). However, this meta-analysis included the New York mortality figures and this trial was not randomised nor were study groups comparable.
The results of the individual trials are also important and a number have now reported later results. At the time of the meta-analysis the Minnesota trial data was complete to 1993, and there was a large reduction in mortality for those who were screened annually (33%) but no significant reduction in mortality among those having biennial screening. However, by 1999, and 18 years of follow-up, the cumulative colorectal mortality rate in the annual group was 33% lower than in the control group (relative risk [RR] 0.67, 95% CI 0.51-0.83) and the biennial screening group also showed a reduction in colorectal mortality of 21% (RR 0.79; 95%CI 0.62-0.97). (Mandel 1999) The reduction in mortality for the Minnesota trial was larger than for the other trials. However, much higher numbers of patients also had colonoscopies (38% and 28% for those screened annually and biennially screened); the participants were from well-motivated groups of volunteers; and re-hydration of the Hemoccult test resulted in a higher sensitivity, and positivity of 9.8% (but increased false positives). After 18 years follow-up, for the first time, there was a reduction in the incidence of colorectal cancer rather than just mortality, with cumulative ratios of colorectal cancer in the annual screening group at 0.80 (95% CI 0.7-0.9), and 0.83 (95% CI 0.73-0.94) for the biennial screening group, compared to control. (Mandel 2000)
In the Nottingham trial, involving 152,850 individuals, biennial screening with unrehydrated Hemoccult tests(and some dietary restrictions) led to reduction in mortality from colorectal cancer in the intervention group of 15% (95% CI; 0.74-0.98) after 7.8 years follow-up (the time of meta-analysis), and 13% (95% CI; 3.0 -22) after 11 years follow-up, compared to control group. However, there was a 27% reduction in mortality (OR 0.73 [CI 0.57-0.90]) at 11 years follow-up, for those who accepted the first hemoccult test (and did not decline testing), compared with the control group. One screening was completed by at least 59.6% of participants, with a positivity of 2.1% by 7.8 years follow-up.(Hardcastle 1996) The cumulative proportion of those who underwent colonoscopy by 11 years follow-up was 1.9%.
In the Funen study, involving 61,933 people, biennial screening with unrehydrated Hemoccult-11 tests (and some dietary restrictions), led to a reduction in mortality from colorectal cancer of 18% (RR 0.82, [CI 0.68-0.99]) after 10 years follow-up (the time of meta-analysis) (Kronborg 1996) and 15% (RR 0.85 [CI 0.73-1.00]) after 13 years follow-up. (Jorgenson 2002) Two thirds accepted the initial screening invite and in those who adhered to the programme the reduction in mortality was at least 30% (RR 0.70) However, those who did not participate in previous rounds were not reinvited unless they had positive findings. The cumulative risk of having a positive test was 5.1% and 4.8% of the study population underwent colonoscopy.
In the non-randomised trial in Burgundy involving 91,199 individuals between 45 and 74 years, biennial screening using unrehydrated Haemoccult (with no dietary restriction) led to an overall colorectal cancer reduction in mortality by 16% (mortality ratio 0.84, CI 0.71-0.99) over 11 years. (Faivre 2004) These results were not available at the time of meta-analysis. Of the screen-detected cancers, 41.8% were at stage Duke’s A. Overall compliance was 55.3%, with 69.5% of the invited population completing at least one test with 1.4% positivity. The positive predictive values of the test was 11.5% for colorectal cancer, 16.8% for large adenomas, and 11.4% for small adenomas. As a result, a screening programme, in people aged 50-74 years will be started, initially in 20% of the population.
In the Swedish trial, involving 68,308 partcipants, initial uptake was 63% with a positivity rate of 4.4%. Uptake dropped to 60% in later rounds. (Kewenter 1994) Mortality data are not available from this study.
The UK Screening Pilot
The UK Colorectal Cancer Screening Pilot was carried out in selected areas of England and Scotland to determine the feasibility of screening for colorectal cancer in the UK using faecal occult blood testing (Hema-screen; biochemically identical to the test used in Nottingham). It involved 478,250 residents, aged 50-69 years, who were sent a test kit by post, to collect six samples. A test was considered weakly positive if 1 to 4 samples were positive and was repeated with dietary restrictions to reduce the risk of reduced sensitivity. Anyone with a positive result was referred for colonoscopy. The early results of this pilot (Steele 2004) have shown that the beneficial results of screening are achievable outside the context of a randomised controlled trial. Screening uptake was similar to the Nottingham trial at 57% overall with a higher response from women (61%) than men (52%). Uptake was lower for younger men aged less that 55years (47%) The overall positivity was 1.9% (England 1.6%, Scotland 2.1%), and increased with age. The positive predictive values of a positive test were 10.9% for invasive cancer and 35% for adenoma. Of 552 cancers detected at screening, 48 % were at Dukes’s stage A, and only 1% had metastases. People aged over 70 years were not offered the test because of a fall off after this age in the Nottingham trial. However, data from the Funen trial suggests that lowering the age of completion of the trial from 74 to 69 years would have missed 25% of the screen detected cancers. (Muto 1975) Over 10% of participants with a positive result for FOB did not take up the colonoscopy. Colonoscopy was complete in 3,700 of 4116 people, a completion rate (full view of colon to caecum) of 89.9% which is higher than the UK average and an indication of the high quality of the examinations. Colonoscopy was associated with some morbidity; ten patients were admitted (0.24%) and 13 patients readmitted (0.32%) with bleeding or abdominal pain (0.24%); and two people had perforations.
Other faecal tests
Other stool tests are being developed which may lead to increased sensitivity or specificity. However they have not been evaluated with respect to mortality reduction. (Pignone 2002)
Screening by flexible sigmoidoscopy
A randomised controlled trial of screening for colorectal neoplasia compared faecal occult blood testing to faecal occult blood testing plus flexible sigmoidocopy (FOS) in 6371 asymptomatic people. (Berry 1997) Adherence with FOB testing alone was 50% and for the FOB testing and sigmoidoscopy combined was 48% for the FOB test but only 20% for FOS. The study measured neoplasia yield, which included both adenomas and cancers, and was four times greater with combined tests although the authors concluded that strategies to improve compliance were required before flexible sigmoidoscopy could become a population screening test.
A randomised controlled trial of 6,367 people between the ages of 55 and 56 years in Sweden, compared rehydrated hemoccult testing to flexible sigmoidoscopy. (Brevinge 1997) Attendance was 59% for Hemoccult; 4% had a positive test and 13% of those had a neoplasm of over 1cm at endoscopy. Attendance for sigmoidoscopy was 49%; and 2.3% had a neoplasm over 1cm. However, using rehydrated Hemoccult, the sensitivity for all neoplasia increased to 26% with a specificity of 95.6%. To find one neoplasm over 1cm, 44 direct sigmoidoscopies were needed whereas in the Hemoccult group only 7 examinations were needed to find one adenoma over 1cm. The authors concluded that the hemoccult test was an alternative to the resource consuming sigmoidoscopy of all invited people.
Although there is indirect evidence from case-control studies there have been no major randomised controlled trials of screening using initial sigmoidoscopy which have yet demonstrated a reduction in mortality.
One small randomised, controlled trial in 799 individuals, assessed colorectal cancer mortality after initial screening with flexible sigmoidoscopy followed by colonoscopic screening if any adenomas were present. Results were of doubtful significance with a relative risk of death of 0.2 (95% CI 0.03-0.95) from bowel cancer, but a significant increase of overall mortality in the screening group. The authors cautioned that a larger trial should look at overall mortality with screening by colonoscopy. (Thiis-Evensen 1999)
Two randomized (volunteer) controlled trials of the effect of a single screening sigmoidoscopy at around age 60 years on colorectal cancer incidence and mortality, are ongoing and have reported baseline results. (Segnan N 2002, Atkins 2002).
In the first study, 34,392 people, who responded positively to a questionnaire about whether they would accept screening (23.9% of those invited), were enrolled. Screenees with colorectal cancer, polyps larger than 5 mm, three or more adenomas, adenomas 5 mm or smaller with a villous component of more than 20%, or severe dysplasia were referred for colonoscopy. In total, 832 people (8.4%) were referred for colonoscopy and another 20 people (0.2%) went directly for surgery. A total of 54 subjects was found to have colorectal cancer, a rate of 5.4 per 1000 (54% of which were Dukes' A). There were two perforations (one in 9911 sigmoidoscopy exams and one in 775 colonoscopies) and one haemorrhage requiring hospitalization after polypectomy during colonoscopy. The pain associated with sigmoidoscopy was described as mild, or less than expected, by 83.3% of the screenees.
Similarly, in the other study, 194,726 people [55%] responded positively to a questionnaire, of whom 170,432 were eligible for randomisation. Small polyps were removed during the screening process and colonoscopy undertaken only in patients with high risk polyps; diameter 1cm or larger; three or more adenomas; tubulovillous or villous histology; severe dysplasia or malignancy; and 20 or more hyperplastic polyps above the distal rectum. (Atkins 2002) On initial screening, 6% of lesions were considered ‘significant’ and these patients were referred for colonoscopy. There were 0.3% cancers including proximal cancers; and in those with follow-up colonoscopy due to high risk lesions, 18.8% patients had proximal adenomas, of whom 0.4% had a proximal cancer; 62% of the cancers were at stage Dukes’ A. However, of the 36 cases of advanced colorectal cancer detected, 10 patients had reported rectal bleeding or increased frequency of bowels prior to the trial (and would normally have been excluded from this type of trial). The overall detection rate for colorectal cancer was 3.5 per 1000 screened. There were 0.002% perforations at flexible sigmoidoscopy and 0.17% at colonoscopy.
One of the concerns around screening with flexible sigmoidoscopy is whether the likelihood of proximal lesions can be predicted. In 2003, a systematic review and meta-analysis of screening colonoscopies, assessed the effectiveness of screening sigmoidoscopy in detecting proximal adenomas and cancers. (Lewis 2003) Thirteen studies met inclusion criteria; of these, ten studies provided data on the prevalence of proximal adenomatous neoplasms regardless of the histology and size of the lesion; and three studies reported data on the prevalence of advanced proximal neoplasms which included cancer, high-grade dysplasia, and villous features in all 3 studies and size greater than 10mm in 2 studies. The meta-analysis pooled estimates of the odds ratio for the association between findings in the distal and proximal colon and the prevalence of isolated proximal adenomatous neoplasia. There was however substantial heterogeneity between studies, partly explained by the investigators definition of the distal colon. A relatively modest association was seen between distal colonic adenomas and the prevalence of adenomas in the proximal colon ([OR] 2.4; 95% CI 1.42-4.05) They also concluded that sigmoidoscopy programmes that referred only patients with distal colonic adenomas for colonoscopy would miss isolated advanced proximal adenomas in approximately 1.3% to 4% of patients.
Screening by Colonoscopy
Colonoscopy is considered to be the gold standard for detecting polyps throughout the bowel. There are, however, no randomised controlled trials assessing colonoscopy as a primary population screening tool, just observational studies. (Winawer 1993, Lieberman 2000). Colonoscopy requires sedation and skilled personnel, is expensive, and has a higher risk of complications,(Pignone 2002) particularly when polypectomy is performed. It is difficult to estimate the risk of harm in a screening population without good randomised controlled trials, but two observational studies have looked at risks. In the first, 3,121 male veterans between the ages of 50 and 75 years, were screened using colonoscopy, of whom 10 (0.3%) had major complications during or immediately after the procedure. (Lieberman 2000) These included six who had bleeding resulting in hospital admission and one with each of the following; stroke; myocardial infarction; Fournier’s gangrene; and thrombophlebitis. In the other study, 1995 patients aged 50 years or over underwent colonoscopy. (Imperiale 2000). One (0.05%) had a perforation and three (0.15%) had bleeding that did not require admission. Most other studies do not distinguish between diagnostic and therapeutic procedures. The US/Preventive Task Force (Berg 2002) has shown that colonoscopy is no more cost-effective than FOBT or flexible sigmoidoscopy. Two Canadian reviews (CTF 2001, McLeod 2001) emphasize the need for randomized controlled trials of this modality of screening before any firm conclusions can be made.
Computed Tomographic (CT) Colonography
CT colonography or virtual colonoscopy is a developing technology being evaluated for colorectal cancer screening. Two recent systematic reviews have highlighted the potential of CT colonography as a screening tool but most of the studies used enriched populations to increase the prevalence of abnormality; the sensitivities were not consistent; and there was heterogeneity between the study results. (Mulhall 2005, Halligan 2005) However, it was noted that the effectiveness of the CT scanning improves as technology advances.
The National Bowel Cancer Screening Programe
Bowel cancer screening will be rolled out across England over the next 3 years, starting in September 2006. It is anticipated that five programme hubs and 14 local screening centres will be operating by Spring 2007 (out of a total of 90 to 100 for full national coverage). (NHS Cancer screening programmes 2006) A similar programme will also be introduced in Scotland. National screening services will need to be of high quality because of the increased demand from screening and surveillance colonoscopies, as well as those for people with symptoms. (Steele 2004) Patients should be advised that screening programme does not detect all cancers, and anyone presenting with symptoms of colorectal cancer, despite screening, should be investigated immediately.
Screening will be by faecal occult-blood testing in those people aged 60 to 69 years. An options appraisal was done for the English Bowel Cancer Screening Group in 2004 which evaluated a number of options, including screening people age 50 to 69 years, in line with the pilot, and a number of ‘once-only’ flexible sigmoidoscopy screening options.(Tappenden 2004) All options were expected to produce cost gains at a cost considered acceptable to the NHS. The most costly option was the FOB screening from 50 to 69 years which reflects the pilot study most closely. It suggested that if endoscopy services were constrained that the favoured option would be FOB screening from age 60 to 69 years, however, this option would be least effective in terms of quality adjusted life days saved. Although other options such as sigmoidoscopy might have proved more cost-effective, this is mainly theoretical as there have not yet been the randomised controlled trials to prove their effectiveness in reducing mortality from colorectal cancer.
Population screening for colorectal cancer with faecal occult blood tests has been shown to reduce mortality from colorectal cancer. Biennial guaiac faecal occult-blood testing has been shown to be successful in randomised controlled trials in normal populations. The benefits of direct flexible sigmoidoscopy screening will not be known for another few years, and at the moment there is no trial evidence to determine the effect of population screening with colonoscopy.
The rollout of population screening in England and Scotland, for those aged 60 to 69 years over the next 3 years is welcomed. However, the impact of the extra screening on demand for endoscopy will be large, and services must be adequately resourced.
Adlercreutz H et al. Five guaiac-based tests for occult blood in faeces compared in vitro and in vivo. Scand J Clin Lab Invest 1984; 44: 519-28.
Simon JB. Occult blood screening for colorectal carcinoma: a critical review. Gastroenterology 1985; 88: 820-37.
Alquist DA. Fecal occult blood testing for colorectal cancer: can we afford to do this? Gastroenterol Clin N Am 1997; 26: 41-55.
Anderson ML et al. Endoscopic perforation of the colon: lessons from a 10-year study. Am J Gastroenterol 2000; 95: 3418-22.
Atkin WS et al. Long-term risk of colorectal cancer after excision of rectosigmoid adenomas. N Engl J Med 1992; 326: 658-62.
Atkin WS et al. Prevention of colorectal cancer by once-only sigmoidoscopy. Lancet 1993; 341: 736-40.
Atkin WS et al. Single blind, randomised trial of efficacy and acceptability of oral picolax versus self administered phosphate enema in bowel preparation for flexible sigmoidoscopy screening. BMJ 2000; 320: 1504-09.
Atkins WS et al. UK Flexible Sigmoidoscopy Screening Trial Investigators. Single flexible sigmoidoscopy screening to prevent colorectal cancer: baseline findings of a UK multicentre randomised trial. Lancet 2002; 359: 1291-300.
Belchetz LA et al. Changing causes of mortality in patients with familial adenomatous polyposis. Dis Colon Rectum 1996; 39: 384-7.
Berry DP et al. Randomised trial of the addition of flexible sigmoidoscopy to faecal occult blood testing for colorectal neoplasia population screening. Br J Surg 1997; 84(9): 1274-6.
Bulow S et al. Centralized registration, prophylactic examination, and treatment results in improved prognosis in familial adenomatous polyposis. Results from the Danish Polyposis Register. Scand J Gastroenterol 1995; 30: 989-93.
Burt RW. Colon cancer screening. Gastroenterology 2000; 119:837-853.
Cairns S, Scholefield JH. Guidelines for colorectal cancer screening in high risk groups. Gut 2002; 51 (Suppl V1-V28)
Colorectal cancer screening. Recommendations statement from the Canadian Task Force on Preventive Health Care. CMAJ. 2001; 165-206-8
Cancer Research UK, June 2006. Cancer Stats - UK [online] Available: http://info.cancerresearchuk.org/cancerstats/types/bowel/ [Accessed 1 June 2006]
Department of Health, Clinical Outcomes Group Cancer Guidance sub-group. Improving Outcomes in Colorectal Cancer; Guidance on Cancer Service. The Manual. I-73. 2003.
Doran J, Hardcastle JD. Bleeding patterns in colorectal cancer: the effect of aspirin and the implications for faecal occult blood testing. Br J Surg 1982; 69: 711-3.
DTB 2006. Bevacizumab and cetuximab for colorectal cancer. DTB 2006; 44: 37-40.
Eguchi S et al. Mutations of the p53 gene in stool of patients with resectable colorectal cancer. Cancer 1996; 77 (8 Suppl): 1707-10.
Faivre J et al. Reduction in colorectal cancer mortality by fecal occult blood screening in a French controlled study. Gastroenterology. 2004; 126: 1674-80.
Farrands PA, Hardcastle JD. Accuracy of occult blood tests over a six-day period. Clin Oncol 1983; 9: 217-25.
Faux W et al. The prevalence and size of adenomatous polyps increases with age. Colorectal Dis. 2002; 4 (Supple 1): 45.P013
Feinleib M, Zelen M. Some pitfalls in the evaluation of screening programs. Arch Environ Health 1969; 19: 412-15.
Gilbertsen VA et al. The design of a study to assess occult-blood screening for colon cancer. J Chron Dis 1980; 33: 107-14.
Hardcastle JD et al. Randomised control trial of faecal-occult-blood screening for colorectal cancer. Lancet 1996; 348: 1472-7.
Herzog P et al. Fecal blood loss in patients with colonic polyps: a comparison of measurements with 51 chromium-labeled erythrocytes and with the Haemoccult test. Gastroenterology 1982; 83: 957-62.
Hutchison GB. Evaluation of preventive services. J Chronic Dis 1960; 11: 497-508.
Imperiale TF et al. Risk of advanced proximal neoplasms in asymptomatic adults according to the distal colorectal findings. N Engl J Med 2000, 343: 169-174.
Jarvinen HJ et al. Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer. Gastroenterology 2000; 118: 829-34.
Jorgensen OD et al. A randomised study of screening for colorectal cancer using faecal occult blood testing: results after 13 years and seven biennial screening rounds. Gut 2002; 50: 29-32.
Kewenter J et al. Results of screening, rescreening, and follow-up in a prospective randomized study for detection of colorectal cancer by fecal occult blood testing. Results for 63,308 subjects. Scand Journal Gastroenterol 1994; 29: 468-73.
Konishi F, Morson BC. Pathology of colorectal adenomas: a colonoscopic survery. J Clin Pathol 1982; 35: 830-41.
Kronborg O et al. Randomised study of screening for colorectal cancer with faecal occult blood test. Lancet 1996; 348: 1467-71.
Kronborg O et al. Randomised study of screening for colorectal cancer with faecal-occult-blood test. Lancet 1996; 348: 1467-1471.
Levin TR, Palitz AM. Flexible sigmoidoscopy: an important screening option for average-risk individuals. Gastrointest Endosc Clin N Am 2002 12: 23-40.
Lewis JD et al. Detection of proximal adenomatous polyps with screening sigmoidoscopy: a systematic review and meta-analysis of screening colonoscopy. Arch Intern Med 2003; 163(4): 413-20.
Lieberman DA et al. Veterans Affairs Cooperative Study Group 380. Use of colonoscopy to screen asymptomatic adults for colorectal cancer. N Engl J Med 2000; 343:162-168. (erratum appears in N Engl J Med 2000; 343: 1204).
Loeve F et al. Endoscopic colorectal cancer screening: a cost-saving analysis. J Natl Cancer Inst 2000; 92: 557-563.
Mandel JS et al. Colorectal cancer mortality: effectiveness of biennial screening for fecal occult blood. J Natl Cancer Inst 1999; 91: 434-7.
Mandel JS et al. The effect of fecal occult-blood screening on the incidence of colorectal cancer. N Engl J Med 2000; 343: 1603-7.
McLeod RS. Screening strategies for colorectal cancer: a systematic review of the evidence. Can J Gastroenterol 2001; 15: 647-60
Morson BC et al. Histopathology and prognosis of malignant colorectal polyps treated by endoscopic polypectomy. Gut 1984; 25: 437-44.
Mulhall BP et al. Meta-analysis: Computed tomographic colonography. Ann Intern Med 2005; 142: 635-50.
Muller (a) AD, Sonnenberg A. Prevention of colorectal cancer by flexible endoscopy and polypectomy. A case-controlled study of 32,702 veterans. Ann Intern Med 1995; 123: 904-10.
Muller (b) AD, Sonnenberg A. Protection by endoscopy against death from colorectal cancer: a case-controlled study among veterans. Arch Intern Med 1995; 155: 1741-8.
Muto T et al. The evolution of cancer of the colon and rectum. Cancer 1975; 36: 2251-70.
Newcomb PA et al. Long-term efficacy of sigmoidoscopy in the reduction of colorectal cancer incidence. J Natl Cancer Inst 2003: 95: 622-5
NHS Cancer screening programmes 2006. Press release. [Online] Accessed on 19th April 2006. http://www.cancerscreening.nhs.uk/bowel/news/005.html
NHS Executive. Guidance on Commissioning Cancer Services: Improving Outcomes in Colorectal Cancer - the manual. Leeds; NHS Executive. 1997. Chairman of the Cancer Guidance sub-group of the Clinical Outcomes Group.Professor RA Haward. (Website: http://www.dh.gov.uk/assetRoot/04/08/02/83/04080283.pdf or http://www.dh.gov.uk/PublicationsAndStatistics/Publications/PublicationsPolicyAndGuidance/PublicationCONTENT_ID=4006658&chk=ysR/7Q
Nivatvongs S et al. Distribution of large bowel cancers detected by occult blood test in asymptomatic patients. Dis Colon Rectum 1982; 25: 420-1.
Office for National Statistics. Cancer statistics registrations. Registrations of cancer diagnosed in 2003. London. HMSO 2005.
Parry S et al. Members of the National Health Committee Working Party on Population Screening for Colorectal Cancer: Recommendations on population screening for colorectal cancer in New Zealand. NZ Med J. 1999; 112: 4-6.
Pignone MP et al. Screening for Colorectal Cancer in Adults. Guide to Clinical Preventive Services, 3rd Ed. Evidence syntheses [online] Accessed on 19.7.2006 at www.ncbi.nlm.nih.gov/books/bv.fcgi?rid=hstat3.chapter.2...
Podolosky DK. Going the distance - the case for true colorectal-cancer screening [editorial] N Eng J Med 2000; 343: 207-208.
Pignone M etc al. Meta-analysis of dietary restriction during fecal occult blood testing. Eff Clinl Pract 2001: 4; 150-6.
Sackett DL et al. Clinical Epidemiology - a basic science for clinical medicine. 2nd Edition Chapter 5: Early diagnosis; pages153-170. Boston. Little Brown and Co. 1991
SEER. Cancer stat fact sheets – Cancer of the Colon and Rectum. [online] Accessed on 24.7..2006 at http://seer.cancer.gov/statfacts/html/colorect.html
Selby JV et al. A case-control study of screening sigmoidoscopy and mortality from colorectal cancer. N Engl J Med 1992; 326: 653-657.
Scottish Intercollegiate Guidelines Network. Management of Colorectal Cancer. 2003; 67. [online] Accessed on 17.7.2006.
Steele RJ et al for the UK Colorectal Cancer Screening Pilot Group. Results of the first round of a demonstration pilot of screening for colorectal cancer in the United Kingdom. BMJ 2004; 329: 133-5
Swaroop VS, Larson MV. Colonoscopy as a screening test for colorectal cancer in average-risk individuals. Mayo Clin Proc 2002; 77: 951-956.
Tappenden P et al. Colorectal cancer screening options appraisal. Cost-effectiveness, cost-utility and resource impact of alternative screening options for colorectal cancer. Report to the English Bowel Cancer Screening Working Group. 2004. ScHARR.
Thiis-Evensen E et al. Population-based surveillance by colonoscopy: effect on the incidence of colorectal cancer. Telemark Polyp Study 1. Scand J Gastroenterol 1999; 34: 414-420.
Towler BP et al. Screening for colorectal cancer using the faecal occult blood test, Hemoccult. The Cochrane Databse of Systematic Reviews 1998, Issue 2. Art. No.: CD001216. DOI: 10.1002/14651858.CD001216.
Wagner JL et al. Cost-effectiveness of colorectal cancer screening in average-risk adults. In: Young GP, Levin B, Rozen P, eds. Prevention and Early Detection of Colorectal Cancer. London: WB Saunders Co Ltd; 1996: 321-356.
Winawer SJ et al. Prevention of colorectal cancer by colonoscopic polypectomy. N Engl J Med 1993; 329: 1977-1981.