This study examined the effects of a cardiovascular screening programme on a group of 79 patients in an Edinburgh general practice. Serum cholesterol levels were measured two to four months after the original screening interview and risk scores recalculated to see if the collection of data was reliable; a control group was used. Patient attitudes to this screening were assessed by questionnaire and the repeatability of the risk score calculation investigated. The programme was popular with patients and was successful in changing their reported behaviour. However, it may have been counter-productive in some patients by sanctioning personal habits detrimental to health. This effect was particularly marked in patients at lower risk levels who it is shown may have raised serum cholesterol levels. It is suggested that the risk score calculation could be improved, in particular by improving the accuracy of the serum cholesterol measurement. It is concluded that more research is needed into the behavioural aspects of cardiovascular screening.