To investigate the reliability of home blood pressure (HBP) measurement as it is generally used in practice, we studied 50 consecutive newly referred hypertensive patients that were measuring HBP for at least 6 months before referral. No specific instructions on the technique of HBP measurement were given to the patients. HBP was compared to clinic BP (CBP) by using daytime ambulatory BP (ABP) as a reference method. The technique of HBP measurement used by the patients was assessed on the basis of a detailed questionnaire and the accuracy of the devices was tested against a mercury column. There was no difference between the degree of similarity of ABP with HBP or CBP (mean value of discrepancies and correlation coefficients) for systolic BP (SBP) and little difference for diastolic BP (DBP). However, the variability of discrepancies between HBP and ABP was equal to the corresponding variability between CBP and ABP, indicating that HBP was not better predictor of ABP than CBP. Although patients measured HBP more frequently than generally recommended, poor standardisation of measurement and wrong technique were particularly common. In conclusion, in the present study HBP measurement offered no advantage over CBP in predicting ABP. The theoretical advantages of HBP may be partially offset by incorrect technique and less standardised conditions of measurement. Without patients education and regular maintenance of devices HBP offers no advantage over CBP and should not be used. (This article is based on data presented at the 7th European Hypertension Society Meeting in Milan, June 1995).