Oral cancer and precancer appear to fulfil many of the criteria for a disease suitable for mass screening. Several commercial organisations in the UK have introduced screening for their employees. One program has been formally evaluated over the course of 1 yr. Of 553 company headquarters staff aged > or = 40 yr, 292 (53%) responded to the well-publicised screening invitation and received a simple clinical examination of the oral mucosa from one of two company dentists. In addition, 17 staff were screened from a separate company work-site. After screening, subjects were examined independently by an oral medicine specialist with access to the relevant diagnostic aids. The dentists' screening decisions were validated against the specialist's definitive diagnoses (the 'gold standard'). The true prevalence of subjects with lesions diagnosed as positive (white patch, red patch or ulcer of greater than 2 weeks' duration) was 17 (5.5%). Overall, sensitivity was 0.71 and specificity, 0.99. The compliance rate to screening among headquarters subjects in seven occupational categories did not differ significantly from the occupational profile for all headquarters personnel. Estimates of relative risk of a positive diagnosis were calculated by logistic regression for five independent variables; gender, age, moderate smoking, heavy smoking, and smoking combined with greater than low risk alcohol consumption. Only heavy smoking (> or = 20 cigarettes per day) produced a significant odds ratio (3.43, P < 0.05).