Our aim was to study the relative importance of the factors associated with stable high blood pressure (BP) in adolescents. Out of 17 634 children born in 1964 and screened in 1977, two samples, an upper group [the upper 5% of the distribution curves for both systolic (SBP) and diastolic (DBP) blood pressures] and a lower group (a 10% random sample from the remainder), were chosen for yearly follow-up (1680 boys and 1143 girls) from 1978 to 1981. A multiple logistic function was used to analyse the factors associated with stable high SBP and DBP (with at least three different measurements of BP at or above the 95th percentile). Different sets of variables were studied. Of the boys with stable high SBP, 55% were found in the upper quintile of risk distribution when Quetelet's index was taken into consideration, 63% when Quetelet's index + age + heart rate+sexual maturation, 67% when Quetelet's index + age + heart rate + serum uric acid and 74% when Quetelet's index + age + heart rate + cardiothoracic index + heart volume index were taken into consideration. Of the girls with a stable high SBP, 68% were found in the upper quintile of risk distribution when any of the following variables was added to Quetelet's index + age + heart rate: R and S waves, serum cholesterol or post-load blood sugar. Similar but smaller effects of associated factors on DBP were observed. Our study shows that Quetelet's index and heart rate are the most important factors associated with stable high BP in adolescents.