The suggestion that treating blood pressure to below a certain level may increase IHD mortality is controversial. We investigated the influence of treated blood pressure on mortality in the DHSS Hypertension Care Computer Project. Mortality was examined by quintiles of treated diastolic blood pressure (DBP) in 2,145 patients treated for a minimum period of one year and subsequently followed for an average of four years. One hundred and seventy five patients died; 71 from IHD. In men and women all cause mortality increased with level of treated DBP. In men IHD mortality showed a U-shaped distribution with an age-adjusted rate of 15.2 per 1,000 person years in the lowest fifth (DBP less than 86 mmHg) comparable to that of 15.6 per 1,000 in the upper (DBP greater than or equal to 103 mmHg). A similar pattern could not be established in women due to very few IHD deaths. IHD mortality was further examined separately for men by prior history of IHD. An increase in IHD deaths in the lowest fifth of treated blood pressure was found for men both with and without a history of IHD. No similar pattern of IHD mortality was obtained for untreated DBP or treated systolic pressure. However, we cannot exclude the possibility that the risk of low treated DBP is secondary to ischaemic heart disease.