An analysis of 102 open-access gastroscopy requests from one general practice over 38 months showed that the detection rate of abnormalities was 58%. Even though no predetermined investigation criteria were used these results compare favourably with gastroscopy findings generally and are superior to the detection rate of lesions using barium meals. Only 12% of the patients who underwent gastroscopy required subsequent referral to a consultant. This represents a major benefit, hitherto undocumented, of an open-access gastroscopy service. Considerations of accuracy, safety and cost effectiveness coupled with the availability of efficacious drugs appear to favour the case for open-access gastroscopy for general practitioners.