A group of 49 male patients suffering from severe depressions, have been investigated to determine the factors related to outcome after treatment by E.C.T. By means of a rating scale for assessing the clinical condition, the relationships can be expressed in the form of correlation coefficients. The highest correlations are Length of Illness .46, Paranoid Symptoms .32, Weight −.31, Body-Index .29, Initial Score on the rating scale .28. These are all statistically significant, although it has been shown that the correlations under-estimate the relationship. Other correlations of interest (but not significant) are the “Drop” in blood pressure in the Mecholyl test .26, and the Hobson score .18. A number of other variables were found to have non significant relationships. They include a family history of mental illness, neurotic traits in childhood, or adult life, psychological or physical precipitation of illness, sudden or gradual onset, fluctuating course, previous attacks, depersonalization, diurnal variation, age and blood pressure.
Five variables give a multiple correlation of .62. The significance of these results and the problems of their interpretation are discussed.