Abstract
BACKGROUND: Rectal bleeding is a common symptom. The ability to distinguish those patients having serious underlying pathology from those with self-limiting conditions is a continuing dilemma in general practice.AIM: To determine the factors affecting the predictive and diagnostic value of rectal bleeding for bowel cancer in primary care.DESIGN OF STUDY: One-year prospective observational study.SETTING: Three large general practices.METHOD: Three hundred and nineteen consecutive patients over the age of 34 years consulting their GPs with rectal bleeding were included in the study. Investigation was by flexible sigmoidoscopy or a questionnaire and review of all patients took place after 18 months. The main outcome measures were consultation rates; the prevalence of cancer, colitis and significant polyps in patients presenting with rectal bleeding; its diagnostic value when occurring with or without a change in bowel habit, perianal symptoms and abdominal pain.RESULTS: The consultation rate for rectal bleeding in patients over the age of 34 years was 15 per 1000 per year; 3.4% had colorectal cancer. The prevalence of cancer increased to 9.2% when the rectal bleeding was associated with a change in bowel habit, and to 11.1% when it was without perianal symptoms. Thirty-six per cent of cancer patients had a palpable rectal mass.CONCLUSION: Over 96% of the patients who present to their GPs with rectal bleeding do not have cancer. Greater awareness of the diagnostic value of the different symptom combinations of rectal bleeding could help GPs adopt different management strategies for patients at higher and very low risk of cancer.