French colorectal cancer screening pilot programme: results of the first round

Abstract

Objectives

In France, a national pilot population-based screening programme on colorectal cancer was set up in 2002. In 2006, 23 French districts were included, targeting a population of more than five million people. This programme offers biennial screening using the fecal occult blood test (FOBT) to average risk subjects aged between 50 and 74 years. People receive a letter inviting them to consult their GPs, who in turn provide the FOBT. People with a positive test result are proposed a full colonoscopy. The results of the programme's first-round performance indicators are presented.

Methods

The monitoring centre collected data from GPs and gastroenterologists on follow-up and colonoscopy results for people who were screened positive. Data were transferred to the French Institute for Public Health Surveillance (InVS) for the analysis.

Results

The overall participation rate for the 19 districts having completed a first screening round reached 42%. The overall positive test rate was 2.7%. In the eight districts with comprehensive follow-up data for the first round, 86% of colonoscopies were completed after a positive test. A total of 1615 people were diagnosed with cancer, and 4612 people with adenoma. The cancer detection rate reached 2.3%, it was higher in men than in women (3.4% versus 1.4%), and increased with age. Forty-three percent of invasive detected cancers were stage I, 24% stage II, 23% had lymph node involvement and 10% presented with distant metastasis.

Conclusion

These results suggest that indicators are consistent with international references. During 2007–2008, the programme coverage will be progressively extended, and all 99 French districts should be actively involved in its implementation.

INTRODUCTION

In France, colorectal cancer (CRC) is a major health problem with 37,400 new cases reported in 2005, ranking second among cancers in women and third among men.¹ Between 1980 and 2000, the number of new cases of CRC increased approximately by 50%. Population ageing explains a large part of these variations.² The highest incidence and mortality rates in Europe were noted in Eastern France.³ In 2004, 16,458 deaths due to CRC were observed in France.⁴ CRC survival was one of the best in Europe, and the overall five-year relative survival rate was estimated at 56%.^5,6 Prognosis is generally better, and treatment more successful if cancer is detected at an early stage. To reduce CRC mortality, prevention and screening are major public health challenges. CRCs are usually preceded by adenoma, which have slow growth rates. The aim of screening is also to prevent CRC, by detecting and removing adenoma. Several randomized population-based studies have demonstrated that screening for CRC using FOBT reduces CRC mortality rates (England, Denmark, USA, Sweden), ranging from 12% to 16% after more than 10-years of follow-up.^7–10 In Bourgogne (France), after 11 years follow-up of 91,200 subjects included in a controlled trial, a reduction of 16% in CRC mortality was observed.¹¹ In France, consensus conferences guidelines were established in 1998 and in 2005 (Haute Autorité de Santé – HAS).¹² A pilot population-based biennial screening programme, using a national protocol published in 2006, has been progressively implemented since 2002 for people aged 50–74.¹³ By 2006, 23 districts (out of 99) were included in the programme. The French Institute for Public Health Surveillance (Institut de Veille Sanitaire [InVS]) is an independent Public Health Agency in-charge of epidemiologic evaluation of cancer screening programmes, in particular the efficacy of the CRC screening pilot programme. The Health Ministry and The French National Cancer Institute coordinate national policy against cancer. Cancer screening programmes are financed by Health Ministry and national health insurance. The objectives of this work were to define and calculate impact, quality and efficacy indicators for the first round of the French pilot screening programme.

POPULATION AND METHODS

Pilot programme in France

The implementation of the pilot screening programme relies on 23 local monitoring centres, which operate at the district level (one per district). These centres are non-profit organizations in charge of GPs training, invitation of the target population, follow-up of screening participants and data collection. They report annual statistics and indicators to the central authorities in charge of epidemiologic evaluation as part of the quality assurance protocol.

Invitation process (invitation letters, recall letters, reminder in case of a positive test not followed by a colonoscopy, result letters for patients and GPs) is based on nominative lists of assured supplied by local health insurance organisms. In France only 0.4% of people do not have health insurance. An information letter is sent by each monitoring centre to men and women aged 50–74, inviting them to consult their GP and collect the free test. The whole target population should be invited by the monitoring centre within two years. If no test was performed, after six months, a reminder was sent by postal mail together with a test. A large number of medical professionals are involved in the programme: GPs, gastroenterologists, pharmacists, anatomocytopathologists. GPs' training plays a major role in the programme.¹⁴ The monitoring centre had to inform and organize local training sessions (on screening, epidemiology, CRC) for all GPs in its area before sending the invitation letter to the population.

General practitioners

The key role of the GPs is to identify persons who should be excluded for medical reasons, explaining and giving the test, explaining the whole screening process (test, colonoscopy, etc.). They also orientate the patient when the test is positive (colonoscopy, treatment, follow-up). The criteria for medical exclusion are:

persons with newly appeared digestive symptoms: these persons should be directly offered a diagnostic colonoscopy

persons at high or very high risk of CRC (FAP; HNPCC, personal or familial history of adenoma or CRC)

persons with a normal colonoscopy during the last 5 years

persons for whom a colonoscopy is considered irrelevant, definitely (e.g. terminal disease) or temporarily (e.g. depression)

GPs have to follow a training session, organized by monitoring centres, to be able to enrol the programme. They eventually receive individualized training. A fee is paid to GPs for each test performed. There were 20,000 GPs potentially involved in the pilot programme, who proposed the test to their patients. Some GPs refused to participate. More than 80% of GPs are involved in the programme (i.e. one test or more given per year). The Guaiac test, Hemoccult II^® (Beckman Coulter, Fullerton, CA), was used and sent to a central reading laboratory after completion at home. There were standardized laboratory quality assurance procedures for central reading laboratories. No diet restriction was recommended. The definition for a positive Hemoccult^® test in the French programme is also less restrictive than in some studies: one or more positive fields out of six is enough to call a test positive. A poor quality FOBT is defined as a technically inadequate test. In case of a technically inadequate test, a new test is sent by mail. In case of a positive test result, the subject is referred to a gastroenterologist by his GP, to undergo a colonoscopy. The French College of Gastroenterology stated that they would provide high quality colonoscopy services, and manage the increase due to the roll out of the programme.

Data collection and indicators

Data were collected by the cancer screening monitoring centres from GPs and gastroenterologists, regarding follow-up and colonoscopy for people screened positive. GPs provided information about ‘medical exclusion criteria’ to the monitoring centre; central reading laboratories provided information about test result to the GP and to the monitoring centre and gastroenterologists transmit information on colonoscopy results to the GP and the monitoring centre. Last but not least the monitoring centre has to gather data on uptake, tests, lesions and key indicators needed for the monitoring and epidemiologic evaluation of the programme and transmit them for national evaluation. This is part of the quality assurance protocol as defined in the list of specifications. One of the agencies that collects statistics from the coordination centre is the InVS (National public health institute for disease surveillance). A standardized form on positive test follow-up have been published in the National Guideline. This form describes colonoscopies quality, possible colonoscopies complication rate, death, surgery, etc. Depending on the starting date and the method of data collection, some districts lacked information at the time of the study. Data collected included gender, age, test results, colonoscopy results and histological results in case of polyps or cancers, the TNM classification was used for tumours.¹⁵ Data on colonoscopy adverse effects (perforation, severe bleeding requiring hospitalization, death) were requested although they were not comprehensive at the time of the study. The participation rate was defined as the number of screened people in the 50–74 years age group divided by the target population. The target population is estimated by subtracting medical exclusions notified to the monitoring centre to the overall population aged 50–74, provided by The French National Institute for Statistics and Economic Studies (INSEE).

RESULTS

Target population

Figure 1 details results of the pilot programme. In 2004, the target population in the 50–74 years age group from the 23 districts amounted to 4,742,525 persons: 2,192,427 men and 2,385,582 women (Table 1). The target population varied by district from a factor of 1 to 8 in the number of people to invite, and the number of GPs potentially involved from 1 to 15. Four districts had not yet completed the first round. They were 19 districts with more than 24 months but less than 36 months from the beginning of the programme and only eight districts with 36 months from the beginning of screening programme.

Figure 1

Flow chart

Table 1

French colorectal cancer pilot programme and participation rate by district – 19 districts

Districts	Beginning of programme	Average target population 50–74 years	Participation rate (%)
Allier	February 2004	104,682	50.4
Bouches-du-rhône	December 2002	487,462	43.7
Calvados	June 2004	20,532	41.0
Charente	September 2003	98,870	32.1
Côte D'or	February 2003	129,260	50.2
Finistère	September 2004	238,921	47.7
Hérault	October 2003	257,880	32.0
Ille-et-Vilaine	December 2002	217,152	50.1
Indre-et-Loire	October 2003	149,623	42.8
Isère	May 2002	277,727	31.1
Mayenne	November 2003	75,837	51.1
Moselle	September 2004	268,077	32.7
Nord	December 2003	593,786	32.2
Orne	June 2004	82,972	41.0
Puy-de-Dôme	May 2004	171,516	37.3
Pyrénées-Orientales	May 2004	123,842	33.8
Haut-Rhin	September 2003	186,200	54.1
Saône-et-Loire	January 2003	158,455	50.3
Essonne	November 2003	276,789	35.1
All		4,742,525	42%

Two years are necessary to invite people and to send first and second reminder letters. Twelve additional months are needed to reclaim and capture data from gastroenterologists about colonoscopy in the monitoring centre's database. The colonoscopy results presented are those, which were completed within the year following the positive test result, and were comprehensive in only eight districts. As shown in Table 1, the overall participation rate reached 42% (1,093,171 participants) and reached more than 50% in only six districts. The exclusion rate for medical reasons varied from 5% to 15%, according to the districts.

FOBT – 19 districts

Among the total number of tests (1,118,251), 3.3% (36,902) were technically inadequate. The rate of technically inadequate tests varied between gender from 3.2% in men to 3.4% in women, and increased with age from 2.7% (50–74 years) to 4.1% (70–74 years) (Table 2). In case of a technically inadequate test, people immediately received an invitation by post including a new test, and an overall 75% rate of people repeated the test. The mean proportion of people with a positive test result was 2.7%, ranging from 2.0% to 3.6% according to district. It was higher in men (3.3%) than in women (2.2%), and increased with age from 2.2% to 3.3%.

Table 2

FOBT results according to gender and age – 19 districts

		Technically inadequate test
	Number of tests performed	Rate (%)	Repeat test rate (%)	Postive test rate (%)
All	1,118,251	3.3	74.8	2.7
Men	491,109	3.2	73.7	3.3
Women	627,142	3.4	76.2	2.2
50–54 years	267,361	2.7	71.4	2.2
55–59 years	259,238	2.9	74.8	2.3
60–64 years	209,369	3.3	77.6	2.8
65–69 years	200,770	3.6	73.7	3.0
70–74 years	180,455	4.1	75.3	3.3

Colonoscopy – eight districts

In the eight districts with more than 36 months after the beginning, among people with positive tests, colonoscopy rate reached 86.2%, and varied from 85.9% in men to 86.5% in women. In the remaining 14%, rates for refusing colonoscopy, for medical contraindication or lost to follow-up were unknown.

Results were available for a total of 14,696 colonoscopies performed (Table 3). In four out of 10 cases, a cancer or one or more adenomas were detected by colonoscopy. This percentage was higher in men than in women (51.1% vs. 29.8%), and increased with age, from 31.5% in people aged 50–54 to 51.2% in people aged 70–74 (not presented). On average, the percentage of cancer detected by colonoscopy was 9.2% (8.0–10.4%), and the percentage of high-risk adenomas (adenomas larger than 10 mm) reached 16.2% (12.0–21.3%). The overall percentage of adenomas detected by colonoscopy was 31.1–22.1% in women and 38.8% in men. A total of 4612 subjects were diagnosed with adenomas. The adenoma detection rate reached 7.2% (4.2% in women and 10.9% in men). The overall cancer detection rate was 2.3% (95% CI 2.2–2.4%) for screened people, and varied from 1.4% in women to 3.4% in men. The percentages of cancers or adenomas increased with age. For cancer, the percentage was three-fold higher in the 70–74 age group than in the 50–54 age group. The adenoma detection rates per 1000 screened people increased with age, from 4.8% in the 50–54 years age group to 10.4% in 70–74 years age group. Cancer detection rates increased with age, from 0.8% for people aged 50–54 years to 4.3% in those aged 70–74.

Table 3

Colonoscopy results according to gender and age – eight districts

	Number of colonoscopies performed	Performed colonoscopy rate (%)	% Adenomas detected (%)	% Adenomas ≥ 10 mm (%)	Adenoma rate per 1000 screened people (%)	% Cancer detected (%)	Cancer detected rate per 1000 screened people (%)
All	14,696	86.2	31.1	16.2	7.2	9.2	2.3
Men	7974	85.9	38.8	21.2	10.9	11.5	3.4
Women	6722	86.5	22.1	10.4	4.2	6.7	1.4
50–54 years	2819	85.1	26.2	12.7	4.8	4.5	0.8
55–59 years	3008	86.3	29.4	15.8	6.1	5.4	1.5
60–64 years	2989	86.9	29.4	14.3	7.2	8.6	2.3
65–69 years	2956	86.8	34.0	17.4	8.9	12.2	3.3
70–74 years	2924	85.8	35.9	20.5	10.4	14.3	4.3

Stage at diagnosis

Cancers were classified by stage at diagnosis according to the TNM classification (International Union against Cancer, UICC). Despite national recommendations, some monitoring centres did not use this classification or had incomplete data. Overall, 1615 cancers were detected by the pilot programme, and 1341 (83%) had information available on staging. Among these, 23% were in situ cancers (pTisNOMO) (not presented). Among the 898 invasive cancers with staging, 43% were classified as stage I, 24% as stage II (pT3T4N0M0), 23% as stage III with lymph node involvement and 10% presented with metastasis (Table 4).

Table 4

Stage of invasive CRC detected

	Invasive cancer detected (n = 898) (%)
Stage I – pT1T2N0M0	43
Stage II – pT3T4N0M0	24
Stage III – pN+	23
Stage IV – pM+	10

DISCUSSION

According to the European population-based FOBT trials, and in order to obtain a mortality reduction, the participation rate should reach 50% or more.^9,16 The French programme's first round participation rate remained limited in comparison with other programmes.^17–19 In the French national breast cancer screening programme the participation rate increased gradually, from 33% in 2003 to 40% in 2004, and 45% in 2005, after national coverage was completed and wide range advertising campaigns were implemented.^20,21 The national implementation of the CRC screening programme, scheduled for 2007–2008, will probably have a positive effect on participation, due to advertising and media coverage. More than 80% of GPs are involved in the programme, which means that they give one FOBT or more per year (data not published). Some of them proposed less than 10 FOBTs per year and some of them more than 100 FOBTs per year. During the pilot programme, only local media campaign were promoted: press article, radio campaign, poster in medical practice. Moreover, FADO (Adherence Factors to Organized Screening Programme), a cross-sectional study, was set up between December 2006 and April 2007 in six districts using a self-administered postal questionnaire. The aim of this study was to analyse the role of sociodemographic and health-care access characteristics according to CRC screening practices and to describe the reasons for participation or not in the CRC screening programme (still in process).

The number of medical exclusions subtracted from the denominator of the participation rate, such as colonoscopy performed in the last five years, varied by district. Large variations in the percentage of excluded people between districts were also observed. This is due to the fact that consensual decisions to subtract medical exclusions from denominator were taken in 2005, and therefore data collection systems for medical exclusions varied widely by district before this date.

In terms of programme quality, the percentage of technically inadequate tests was in agreement with the French reference rate (2–3%).^16,22 This underlines the quality of GP training and the relevance of the information given to the invited population, especially on how to perform the test. Test readings are centralized in qualified local laboratories, and test readers are trained by a national reference centre. In addition, the average positive test rate of 2.2% was consistent with the Bourgogne study for a population aged 45–74, where the positive test rate was 2.1% at the first round.¹¹ This rate was observed, although no diet restriction was recommended, to increase with age in all areas, and varied according to gender and the natural history of CRC. In Italy and Spain, age and gender trends were similar.^23,24 In Italy and the UK, positive test rates were also higher in men than in women. One year or more after test results, not all colonoscopy results were recorded by the monitoring centre. The colonoscopy completion rate reached 86%, which was higher than in the UK – 82%, and Italy – 75.3%.^17,23 In the Spanish pilot programme in Catalonia, this rate reached 86.6%.²⁴ In the districts where the percentage is high, specific strategies for data collection from doctors were in place (reminders, letters or mails to physicians, phone calls, and as a last resort, posting a registered mail to the person with a positive test). In the French pilot programme, among the 14% of those who did not undergo colonoscopy, some people refused colonoscopy. However, quantification of the refusal rate was not yet available at the time of writing. In Bourgogne, during the first round, the percentage of people who did not perform colonoscopy was 12.7%.¹¹ People are free to refuse, but this result may also reflect a lack of understanding of the screening phases (test and colonoscopy), or patient's preferences. The rates of cancers or adenomas detected by colonoscopy were similar to those reported in other studies. The percentages varied by country between 10% and 15% for colonoscopies detecting CRC, and between 30% and 40% for one or more adenomas.^11,16 In Bourgogne, the percentage reached 11.4% for cancer.¹¹ For adenomas overall, the percentage reached 20.3% in Italy and 35% in the UK.¹¹ International programmes set the bench-mark to be around 30% for adenomas, and 10% for CRC. During the first round, in Italy, Spain, the UK and Bourgogne cancer detection rate reached 2.8%, 2.1%, 1.63% and 1.7%, respectively.^11,17,23,24 The lower target age group was slightly younger in the Bourgogne trial (45–74 years), the upper target age group, younger in Spain and in the UK (50–69 years). Since cancer detection rate increase with age, the observed age variations may explain differences between these rates.

The overall adenoma detection rate of 7.2% was consistent with the rate found in the UK: 6.9%, but was lower than the rate found in Spain 9.2%. In Italy, adenomas detection rate range between 6.7% for low-risk adenomas (<10 mm), and 11.6% for high-risk adenomas (≥10 mm). In Bourgogne, detection rates reached 11.4% and 16.8%, respectively. Rate of positive test and rate of detected lesions were higher in men than in women, and cancer detection rates were almost three times higher in men than in women.²⁵ For cancers, unknown stage or stage misclassification differed according to districts, and reached 17% overall. In the Italian programme, the percentage of unknown cancer detected stages was 10%. Among invasive cancers, stage I of the TNM classification represented the majority of the cases, as in Bourgogne, of the 196 CRC-screened detected individuals – 43% were stage I.¹¹ In Italy, 33% of detected cancers were stage I.²³ One of the aim of the programme is to detect cancers at an early stage. The first round is generally characterized by high detection rates, since the cancers or lesions detected were prevalent. In the next round, detected lesions should be fewer and present with better prognosis. However, the percentage of CRC with good prognosis observed in the first round of this pilot programme, is higher than the percentage observed in French cancer registries (19%) published in 2007 (from the Francim Cancer Registry Database).

CONCLUSION

In France, men and women aged 50–74 from the 23 pilot districts represent five million people. The results of the pilot CRC screening programme suggested that impact, quality and efficacy indicators were consistent with other pilot programmes and controlled studies in Europe. However, participation rates were still low in some districts. Improvements should be obtained after roll out of the programme, and national advertising and media campaigns, increase the population and GPs awareness for the screening programme. Another essential point is to improve colonoscopy uptake after a positive FOBT, since a high percentage of lesions may be detected. Moreover, there is a need to improve standardized data collection, namely reaching consensus on all major epidemiologic indicators. Some districts still lack comprehensive data related to the follow-up of positive tests, particularly for cancer staging. There is also a need for improving completeness, timeliness and quality of data transmission from GPs, gastroenterologists and pathologists to the monitoring centre, and from the monitoring centres to InVS. In 2005, the French Health Minister decided that national coverage of the programme should be completed. Information on possible adverse reactions of colonoscopy and the stage of cancers are insufficient. Findings from the evaluation of the French Colorectal Cancer Screening Pilot Programme suggest that the National FOBT screening programme is feasible. During 2007–2008, the programme's coverage will be progressively extended, and all 99 districts will be implementing the programme.

Footnotes

ACKNOWLEDGEMENTS

ALLIER (Dr Josette Puvinel), ALPES-MARITIMES (Dr Claire Grannon), ARDENNES (Dr François Maudière), BOUCHES du RHONE (Dr Karim Tifratène), CALVADOS (Dr Véronique Bouvier), CHARENTE (Dr Dominique Mahieu), COTE D'OR (Dr Vincent Dancourt), FINISTERE (Dr Yvon Foll), HERAULT (Dr Jeanine Cherif-Cheik), ILLE ET VILAINE (Dr Gérard Durand – Dr Piette), INDRE ET LOIRE (Dr Jérôme Viguier), ISERE (Dr Catherine Exbrayat), MARNE (Dr Frank Arnold – Dr Michel Thirion), MAYENNE (Dr Pascale Lalanne), MOSELLE (Dr Yves Spyckerelle), NORD (Dr Sabine Chaumette, Dr Brigitte Cockenpot), ORNE (Dr Annick Notari-Lefrançois); PUY DE DÔME (Mme Chantal Mestre); PYRENNEES ORIENTALES (Dr Jean patrice Gautier); HAUT-RHIN (Dr Philippe Perrin); SAÔNE ET LOIRE (Dr Nassime Touillon); ESSONNE (Dr Hamou Ait Hadad – Dr Agnès Bernoux); SEINE ST DENIS (Dr Christophe Debeugny – Dr Samira Saïfi).

References

Remontet

, Esteve

, Bouvier

, Cancer incidence and mortality in France over the period 1978–2000. Rev Epidemiol Sante Publique 2003;51:3–30

Bouvier

, Remontet

, Jougla

, Incidence of gastrointestinal cancers in France. Gastroenterol Clin Biol 2004;28:877–81

Bouvier

, Remontet

, Jougla

, Incidence of gastrointestinal cancers in France. Gastroenterol Clin Biol 2004;28:877–81

INSERM. Centre d'épidémiologie sur les causes médicales de décès causes of death (Cepidc). See http://www.cepidc.vesinet.inserm.fr (last checked October 2007) 2007

Faivre-Finn

, Bouvier-Benhamiche

, Phelip

, Manfredi

, Dancourt

, Faivre

. Colon cancer in France: evidence for improvement in management and survival. Gut 2002;51:60–4

Grosclaude

, Bossard

, Remontet

, , eds. Survie des patients atteints de cancer en France. Paris: Springer 2007: 137–44

Kronborg

, Jorgensen

, Fenger

, Rasmussen

. Randomized study of biennial screening with a faecal occult blood test: results after nine screening rounds. Scand J Gastroenterol 2004;39:846–51

Jorgensen

, Kronborg

, Fenger

. 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

Towler

, Irwig

, Glasziou

, Weller

, Kewenter

. Screening for colorectal cancer using the faecal occult blood test, hemoccult. Cochrane Database Syst Rev 2000;CD001216

10.

Kronberg

, Asplund

, Maki

, Sjoholm

. Reaction of mucochloric and mucobromic acids with adenosine and cytidine: formation of chloro- and bromopropenal derivatives. Chem Res Toxicol 1996;9:1257–63

11.

Faivre

, Dancourt

, Lejeune

, Reduction in colorectal cancer mortality by fecal occult blood screening in a French controlled study. Gastroenterology 2004;126:1674–80

12.

[Consensus Conference: Prevention Screening and Management of the Colonic Cancers. Paris, France, 29–30 January 1998. Proceedings]. Gastroenterol Clin Biol 1998;22:S1–295

13.

Cahiers des charges du dépistage organisé ducancer colorectal. Direction générale de la santé. See http://ile-de-francesantegouvfr/img/pdf/cahier_des_charges.pdf 2006

14.

Denis

, Perrin

, Cailleret

, Guth

, Ruetsch

, Strentz

. [Colorectal cancer screening: a survey of French general practitioners]. Gastroenterol Clin Biol 2003;27:992–7

15.

Sobin

, Fleming

; Union Internationale Contre le Cancer and the American Joint Committee on Cancer. TNM Classification of Malignant Tumors, 5th ed. Cancer 1997;80:1803–4

16.

Faivre

. Dépistage de masse par la recherche d'une saignement occulte dans les selles. Epidémiologie et prévention du cancer colorectal. Paris: Springer 2006

17.

UK Colorectal Cancer Screening Pilot Group. Results of the first round of a demonstration pilot of screening for colorectal cancer in the United Kingdom. Br Med J 2004;329:133

18.

Grazzini

, Castiglione

, Ciabattoni

, Colorectal cancer screening programme by faecal occult blood test in Tuscany: first round results. Eur J Cancer Prev 2004;13:19–26

19.

Tazi

, Faivre

, Dassonville

, Lamour

, Milan

, Durand

. Participation in faecal occult blood screening for colorectal cancer in a well defined French population: results of five screening rounds from1988 to 1996. J Med Screen 1997;4:147–51

20.

Invs. Taux de participation des femmes au programme de dépistage organisé du cancer du sein en France, 2003–2005. See http://www.invssantefr/2006 (last accessed 20 August 2008)

21.

Invs. Taux de participation des femmes au programme de dépistage organisé du cancer du sein en france, 2003–2005. See http://www.invssantefr/2006 (last accessed 20 August 2008)

22.

Levin

, Barthel

, Burt

, Colorectal Cancer Screening Clinical Practice Guidelines. J Natl Compr Canc Netw 2006;4:384–420

23.

Zorzi

, Grazzini

, Senore

, Vettorazzi

. Screening for colorectal cancer in Italy: 2004 survey. Epidemiol Prev 2006;30:41–50

24.

Peris

, Espinas

, Munoz

, Navarro

, Binefa

, Borras

. Lessons learnt from a population-based pilot programme for colorectal cancer screening in Catalonia (Spain). J Med Screen 2007;14:81–6

25.

Launoy

, Herbert

, Vallee

, [Mass screening for colorectal cancer in France. Experience in 165,000 people in the department of Calvados]. Gastroenterol Clin Biol 1996;20:228–36