Objectives: To analyse results of a screening program for colorectal cancer using flexible sigmoidoscopy.
Design: Survey of results of screening program and follow-up colonoscopies and identification of missed cases from State cancer registry data.
Participants: Asymptomatic, average-risk people aged 55-64 years who were either mailed invitations after random selection from the electoral roll or volunteered after hearing about the program.
Setting: Fremantle Hospital, Western Australia (a public teaching hospital), July 1995 to November 1999 (first 4.5 years of the screening program).
Main outcome measures: Participation rates; lesions detected; stage of colorectal cancers diagnosed at the hospital before and after the screening program began (1989-1995 versus 1996-1999); and diagnoses of colorectal cancer in previously screened individuals (from State cancer registry data).
Results: 6446 people were mailed invitations, and 1483 were screened (23% participation rate). Another 1122 people volunteered, giving 2605 people screened overall. Flexible sigmoidoscopy showed adenomatous polyps in 352 people (14%), and colonoscopy was recommended in 399 (15%) on the basis of clinically suspicious lesions. Colonoscopy was performed in 302 (76% participation rate). Screening and follow-up colonoscopy detected 14 colorectal cancers (10 invasive, with eight of these Dukes stage A). One participant was diagnosed with colorectal cancer 12 months after sigmoidoscopy gave normal results. Incidence of colorectal cancer was 119 per 100000 per year, and prevalence was 0.5%. Before the screening program, 12% of cancers diagnosed at our hospital were Dukes stage A, compared with 28% after (P<0.001).
Conclusions: Flexible sigmoidoscopy screening is an acceptable strategy in asymptomatic, average-risk people which detects colorectal cancer and adenomatous polyps. Screening has been associated with a trend to earlier presentation of cancer in our institution.