In a retrospective analysis of open-access versus hospital-referred flexible sigmoidoscopies, the two groups are compared with reference to the demographic data, presenting symptoms, sigmoidoscopy findings and diagnostic yield. Overall, 1090 patients underwent sigmoidoscopy during 12 months, 544 in the open-access and 546 in the hospital-referred group. There was a preponderance of females in both groups, but patients in the hospital-referred group were older. Diarrhoea was the most common presenting symptom, followed by rectal bleeding. Significantly more patients presented with rectal bleeding with or without diarrhoea and abdominal pain in the open-access group, while there were more patients with iron deficiency in the hospital-referred group, The number of patients with colonic carcinoma was similar in the two groups, but significantly more early carcinomas were found in the open-access group. There were significantly more patients with haemorrhoids in the open-access group. The positivity rate was similar in the two groups (52% in the open-access vs 46% in the hospital-referred group), Of the 24% of patients 40 years or under, none had carcinomas. In this age range the positivity rate was no different in the two groups (32% in the open-access vs 23% in the hospital-referred group). The diagnostic yield of open-access flexible sigmoidoscopy is thus comparable to hospital-referred sigmoidoscopy, suggesting that it should be freely available to GPs.