A trial and formulation of regression equations of RV 5 on several constitutional variables including age--relationship between electrocardiogram and body-build.

In a series of healthy subjects ranging in age from 12 to about 20 years, having no cardiac disorders, the interrelationships between the amplitude of the R wave in lead V5 and the various parameters of the body-build including age were clarified as follows: 1. When groups of subjects of comparable age were compared, evidence was obtained for significant differences to exist in these respects between the sexes. 2. Age-related differences in the electrocardiographic deflections are remarkably wide in the male subjects, whereas in the females at 15 to 16 years of age and in those at about 20 years, they were observed to be not significant. 3. Since a high multicollinearity exists among the parameters of the body-build, regression analyses on the relationship between the bodybuild and the electrocardiogram should involve adequate considerations of it. As a means for excluding this multicollinearity, we adopted the procedures of Draper and Smith and presented correction formula with age and chest circumference taken into due account. 4. Among the various parameters of the bodybuild employed, it was found to hold true for both sexes that l/(chest circumference)2 showed apparently greater regression than the chest circumference itself. Account should be taken, therefore, of the chest circumference in the interpretation of the electrocardiogram. 5. A question remains to be unsettled as to whether the amplitude of the R wave in lead V5 itself is suitable or not to be taken as a random variable in the above-mentioned regression equation. 6. The intensity of physical exercise was observed to cause significant differences in the individual parameters of the body-build as well as the amplitude of lead V5, but statistical analyses failed to yield any appropriate correction formulas for them. This requires further consideration in the future.