Beware the nones of March: suicide at the beginning of the month.

Abstract

MacMahon (1983) found that from 1972-1978 suicides peaked on the fifth of the month and were less frequent at the end of the month on the 29th. The purpose of the present study was to see if these results could be replicated and whether they were found for a11 age groups. Data were from the detailed mortality tapes for 1980 available from the National Center for Health Statistics. The number of suicides occurring on the 4th, 5th, and 6th of each month were compared to the number of suicides on the same days of the week one week earlier (a three-day period usually including the 29th and two other days). For the total sample of 26,918 suicides in 1980, there were 137 more suicides on the 4th to 6th of the month than on the comparable periods at the end of the month (t9 = 2.50, one-tailed p < .03). This was true for those aged 35 yr.+, for whom there was an excess of 117 deaths on the 4th to the Gth ( t o = 2.85, p < .01), but not for those aged 15-34 yr. ( t g = 0.50). In an analysis by sex this effect was significant also for the men ( t9 = 2.97, p < .01) in the predicted direction but significant in the opposite direction for women (to = 1.9S, p < .05). On the 4th, 5th, and 6th days of the months for the 11 months there were 1911 male suicides and 524 female suicides. This analysis has confirmed MacMahon's finding and, like Lester and Frank (1987) who found the monthly variation in suicide rates only for those aged 35 yr. and older, the present study shows this phenomenon held for those 35 yr. and older. It is not easy to propose an explanation of this phenomenon at the present time.