Electrocardiographic evidence of left ventricular hypertrophy (ECG-LVH) has a grave prognostic significance in hypertensive patients. The purpose of our study was to assess whether ECG-LVH is more strongly associated with home-measured blood pressure (BP) than with clinic BP, and whether the correlation between home BP and ECG-LVH increases with the number of home measurements performed. We studied a representative sample of the general adult population (1989 subjects 45-74 years of age) in Finland. Subjects included in the study underwent a clinical interview, electrocardiography and measurement of clinic BP (mean of two clinic measurements) and home BP (mean of 14 duplicate home measurements performed during 1 week). Home BP correlated significantly better than clinic BP with the Sokolow-Lyon voltage (home/clinic systolic: r=0.23/0.22, P=0.60; diastolic: r=0.17/0.12, P=0.009), Cornell voltage (systolic: r=0.30/0.25, P=0.004; diastolic: r=0.21/0.12, P<0.001) and Cornell product (systolic: r=0.30/0.24, P=0.001; diastolic r=0.22/0.14, P<0.001) criteria of ECG-LVH. The correlation between home BP and ECG-LVH increased slightly with the number of home measurements, but even the mean of the initial two home BP measurements correlated equally well (systolic BP), or better (diastolic BP) with ECG-LVH than did clinic BP. In conclusion, home BP measurement allows us to obtain a large number of measurements that have a strong association with ECG-LVH. Our data support the application of home BP measurement in clinical practice.