Brain weight in suicide revisited

In their excellent paper Hamilton & McMahon (2002) examined brain weight in suicide victims of all ages to see whether it was higher than in a control group. They attempted to replicate and reinterpret our findings (Salib & Tadros, 2000) reported in an elderly sample. The authors, quite rightly, looked at brain weight in cases and controls adjusted for body mass index (BMI), having collected additional data about body weight and height, data which Salib & Tadros (2000) were not able to collect and which was already accepted as a major limitation in the latter study.

Hamilton & McMahon (2002) did not find any significant difference between brain weight adjusted for BMI in cases and controls. However, brain weight was significantly higher in those dying by hanging than in those dying by overdose.

I would like to make one or two comments which may help to explain the difference in the findings of the two studies. In Hamilton & McMahon's study, the mean age is 38.5 years (for cases and controls) compared with 72 years in the study by Salib & Tadros (2000). Also, the mean brain weight for Hamilton & McMahon's control group was 1449 g compared with 1238 g in the sample reported by Salib & Tadros (2000). Hamilton & McMahon (2002) included only 6% of subjects aged over 60. The method of selection of the control group in their sample is different from that used by Salib & Tadros (2000) — the latter study included only controls who died naturally and not accidentally. Hamilton & McMahon (2002) were not able to replicate our findings in an elderly sample but were careful in their comparison of the findings by taking into account the differences in some basic parameters in the two studies.

It is interesting to note that another recently published paper (Balazic & Marušič, 2002), which included patients of all ages, has confirmed our findings of a higher brain weight in suicide (Salib & Tadros, 2000) but only after they excluded ‘outdoor’ cases (where the body was found at an outdoor location away from the home) and controls. Hamilton & McMahon should be congratulated on their study; however, the negative findings in the younger age group may have been confounded by the choice of the control group, some of whom may have had a mode of death not dissimilar to suicide but had a non-suicide verdict returned by the coroner. On the other hand, Hamilton & McMahon (2002) are correct in making the assumption that our control group may have included people with pre-clinical dementia with lighter brains. This may have had the opposite effect on the findings (i.e. heavier brain weight in elderly suicide cases).