Variations in the mortality with respect to lunar phases

The overall mortality and cardiovascular mortality data from Romania within 1989-1995 were processed. They exhibit an increase by about 10% caused by the increase of inhabitants and by the increase of their average age. The p ercentage of cardiovascular mortality increases as well. Both data series d isplay a pronounced annual wave supplemented by a smaller semiannual wave, resulting in a double maximum in January and April and a minimum in August. There is also a variation with the social week, given by a combination of the 7-day and 3.5-day wave, with a maximum on Thursday and Sunday and a dis tinct minimum on Friday. The mortality depends also on the lunar synodic cy cle. This variation has a form of the semilunar wave with maxima two days b efore first and last quarter. The difference in the mortality between maxim a and minima was tested and appeared to be significant. The semilunar wave is supplemented by some smaller waves with shorter periods but not by a lun ar wave. This semilunar variation agrees surprisingly well with the same va riation found in the sudden cardiovascular mortality data in Brno (400,000 inhabitants), Czech republic, within 1975-1983. This data set includes all cases of sudden death due to cardiovascular defeat at home, etc., not in ho spital (i.e., earlier than medical assistance arrived). The maxima of the s emilunar wave lay two days before the first and two days before the last qu arters and their position depends on the solar cycle. Moreover, waves with periods between 3 and 4 days appear to be significant as well. All these pe riodicities in both data sets (from Romania and from Brno) are only connect ed with the lunar month and are not pronounced if these data are arranged w ith respect to some other effects (e.g., solar rotation).