The uptake of faecal occult blood testing in a workplace based colorectal cancer screening programme was investigated. Altogether 1828 employees aged 41-65 years at a large British industrial company were invited to receive a free faecal occult blood test (Haemoccult). Faecal occult blood tests were completed on three separate days. Patients with positive results were invited to undergo colonoscopy. The number of employees completing kits was measured and differences in compliance according to age, sex, and occupation were tested with a x 2 test. Compliance was 25.4%, and similar in men (25.0%) and women (32.0%, x 2 = 3.0, not significant). In men, compliance was highest in those aged 51-60 years (30.5% x 2 .11.6, p,0.001). Compliance in women aged 41-50 years, 51-60 years, and 61-65 years was similar (Yates's corrected x 2 ,2.08, not significant). Managers returned more kits than clerical and blue collar workers (28.6% v 23.5%, x 2 = 5.6, p,0.02). One percent of tests were positive and one patient had a tubular adenoma. Compliance in employees aged 51-60 years was comparable to that achieved in one-off British general practice programmes, but less than that in the large randomised trial of screening in general practices in Nottinghamshire. Health education of large numbers of people is easier at the workplace than in the community. Future screening must target older employees and those with clerical and blue collar jobs.C olorectal cancer is a common cause of death from malignant disease resulting in over 19 000 annual deaths in Britain 1 and 57 000 in the United States. 2 The overall five year survival rate is only 30% 3 but for those with limited disease at presentation (Dukes's type A tumours) it is 85%. 4 In the commonest method of screening, that is, faecal occult blood testing, small stool samples are analysed for microscopic traces of blood from asymptomatic cancers and polyps. Three large studies of faecal occult blood testing showed more early tumours (Dukes's A type tumours) and adenomatous polyps in the screened than control group, [5][6][7] which lead to a reduction in mortality of 15%, 18%, and 33% respectively.The World Health Organisation has stated that one of the criteria necessary for a successful screening programme is a high compliance.8 A major problem in colorectal cancer screening trials is encouraging people to participate. A poor compliance means few people benefit and the economic costs are high. This problem of low compliance was identified in the Nottingham trial where only 60% of the test group completed at least one test, with just 38% returning all the follow up tests at two yearly intervals.5 Unfortunately, compliance is also a problem in other smaller community trials in British general practice, where fewer than 50% of the population participated.9-15 An alternative approach to delivering screening, which may raise compliance, is to develop workplace based screening programmes. On-site health education at the workplace, the availability of occupational medical and nursing staff...