Abstract

Multiple regression techniques were used to determine the most efficient combination of height, weight and body mass index in the prediction of systolic and diastolic blood pressures for a national sample of 13,723 10-year-old children. In every analysis an adjustment was made for the depth of sphygomomanometer cuff used when taking the blood pressure. The variables which together best predicted the systolic blood pressures in boys were weight, height and (height)2, and in girls weight/(height)2 and height. Diastolic blood pressures were predicted best by weight only for boys and by weight/(height)2 and height for girls. Once these factors had been taken into account there was no difference in blood pressures in those children for whom there were signs of puberty.