Objective : to demonstrate that providing to GPs a list of their patients who are not compliant to colorectal cancer screening will 1) enhance patient participation to screening, and 2) decrease the proportion of cancer diagnosed outside the screening organisation. Design : Randomised controlled study, 3 parallel arms. Enrollment: GPs allocated in the 3 groups of the study will be all GPs 1) from the "Loire-Atlantique and Vendée" geographic area (1300 GPs on average) and 2) who have more than 100 patients in their patient list (based on data provided by the Healthcare Insurance System Services). Patients will be eligible to the study depending on their eligibility to the Faecal Occult Blood Test for colorectal cancer screening : 1) they should be older than 50 and younger than 74; 2) they should not have a personal history of colorectal cancer (or history of adenoma bigger than 1 cm) nor a family history of colorectal cancer. Main outcomes measures : Patient participation to colorectal cancer screening, and number of cancers screened in (versus diagnosed outside) the screening procedure.

Background:

Participation rate for colorectal cancer screening in France is 39%. In a Cochrane review, patient participation rates to colorectal cancer screening varies from 30 to 60%. Policy makers usually consider that these participation rates should be higher, based on the premise that the impact of the screening mainly depends on patient participation. In France, patients at average risk of colorectal cancer are eligible to the Faecal Occult Blood Test for colorectal cancer screening if: 1) they are older than 50 and younger than 74; 2) they have no personal history of colorectal cancer (or history of adenoma bigger than 1 cm) and no family history of colorectal cancer.

Objective:

To demonstrate that providing to GPs a list of their patients who are not compliant to colorectal cancer screening will 1) enhance patient participation to screening, and 2) decrease the proportion of cancer diagnosed outside the screening organisation. Design : Randomised controlled study, 3 parallel arms.

Design:

Randomized controlled study. 3 parallel arms.

Setting:

Primary care setting in France.

Enrollment:

In this study, the intervention will focus on the GP. The statistical unit for the results analysis will be the GP. GPs allocated in the 3 groups of the study will be all GPs 1) from the "Loire-Atlantique and Vendée" geographic area (1300 GPs on average) and 2) who have more than 100 patients in their patient list (based on data provided by the Healthcare Insurance System Services).

Intervention:

GPs allocated to the intervention group (A) will receive a list of the patients who are not compliant to colorectal cancer screening, while GPs allocated to the others groups will not receive the corresponding information.

Randomization:

3 groups : (A) GPs will receive a list of the patients who are not compliant to colorectal cancer screening, (B) GPs will receive a document providing general information on colorectal cancer screening, (C) GPs will not receive any specific document (they continue regular practice).

Primary outcome measure: Patient participation to colorectal cancer screening Secondary Outcome Measures: number of cancers screened in (versus diagnosed outside) the screening procedure.

In this study, the statistical unit is the GP.

Inclusion Criteria:

GPs allocated in the 3 groups of the study will be all GPs

• 1) from the "Loire-Atlantique and Vendée" geographic area (1300 GPs on average)
• 2) who have more than 100 patients in their patient list (based on data provided by the Healthcare Insurance System Services).

Exclusion Criteria:

• GPs who have less than 100 patients in their patient list