Seasonality of Suicide Deaths in Serbia, 1990-2012

Background: In the period from 1990 to 2012, there were 32,855 suicides registered in Serbia, i.e. an average of 1428 deaths per year (18.9 per 100.000 inhabitants). The suicides were the most frequent in the early 1990s, less frequent during the 2000s, and least frequent in the last three years of the observed period (2010-2012 – an average of 1237 suicides per year). Objective: The goal of this paper is to evaluate the seasonal variations of suicides in Serbia in the period of 1990-2012, their changes, as well as the accordance with findings from other countries. Method: In the paper are used "classical" statistical methods of evaluating cyclical variations (c2-test, Edwards' test) as well as some frequently used newer methods (e.g. the peak-low ratio). It also introduces a new indicator of the intensity of monthly variations in suicides (the magnitude of trimester variations of /12/ moving consecutive months index – the MtMV index). Seasonality of suicides is also observed by sex. Results: The results of the research of seasonality confirm that certain cyclical variations in mortality due to suicide are also present in Serbia. Observed by season, suicides are most frequent in spring and summer and less in winter. The cyclical nature was also clearly confirmed by month. Suicides are most frequent in "warm" months (April-August) and far less frequent in "cold" months (December-February). By month, suicides are most frequent in May, while the maximal average number of suicides per day was calculated for June (21 % higher than the average for 1990-2012). December is singled out as the month with the lowest number as well as lowest daily average of suicides (25 % lower than the average). Conclusion: The cyclical nature of suicides is clearly notable during the entire observed period and no significant fluctuations or decrease in the intensity of the variation were observed. Regardless of the method of analysis, the resulting conclusions are identical regarding the cyclical nature of the variations and the temporal characteristics of suicides in Serbia at the end of the 20th and the beginning of the 21st century. At the aggregate level there are no significant differences in seasonal variations of suicides committed by men and women. Changes also developed in parallel and in the same direction for both sexes.