Association between a Full Moon and Completed Suicide

There does not appear to be a reliable and significant association between the phase of the moon and incidence of completed suicide in a northern state.

Abstract

Pokorny (1964) investigated whether a sample of completed suicides in Texas were clustered around the incidence of a full moon rather than a new moon. He found no association for the total sample, or for subgroups by sex and race. It was felt to be of interest to examine this association in a northern state since such correlations may differ in the North and the South. For example, although completed suicides show a seasonal variation in the United States as a whole, they do not in Los Angeles. A listing of all the completed suicides recorded in Erie County from 1964 to 1968 was 0btained.l The sex of each suicide was identified and the date of the suicidal act. The number occurring in the 7 days centered around the full moon was compared with the number expected by chance. Of the 399 completed suicides, 102 occurred during the weeks surrounding the full moon; the expected number was 93.25. (Of the 1827 days in the 5-yr. period, 427 constituted those around the time of the full moon.) The difference, though in the direction of a positive association, did not reach statistical significance (x2 = 1.07, df = 1). The data were examined for males and females separately and the association, though positive for both sexes, did not reach statistical significance (for males x2 = 0.22 and for females xk 1.41, df = 1). The association for the total sample was significant for 1967 only (this was produced by the presence of a strong association for the females). However, the association was positive for only 4 of the 5 yr. in the period examined. Thus, the present data appear to confirm those of Pokorny. There does not appear to be a reliable and significant association between the phase of the moon and incidence of completed suicide.