[Correlation between weather factors and the number of premature and yearly births].

In a retrospective study for a period of five years, a total of 41,760 births in Sarajevo were analysed. For the same period, also meteorological data, such as air temperature, humidity, rain, cloudiness, barometric pressure, were gathered, by determining the weather type for each day. Using computer analyses, daily weather types with the number of births were compared. On the basis the statistical method of Friedman's test it is concluded that there is a significant correlation (p less than 0.05) between the weather types and the daily number of births. By comparing the daily number of births with the weather type of the preceding day the same conclusion was drawn. A cyclic fluctuation was also observed in the number of human births: the increased number of births in the December-January period and the decreased number of births in the August-September period. Premature births did not show any cyclic pattern.