Mortality from congenital anomalies.

Research has been conducted on the relationship between postnatal mortality from dominant causes and age, using data from the US, Japan and former Czechoslovakia, during the period 1979–1991. The logarithm of a total mortality, total mortality excluding deaths caused by accidents, logarithm of mortality caused by congenital anomalies, logarithm of mortality caused by other respiratory diseases and logarithm of mortality caused by diseases of the nervous system fell linearly with the logarithm of age, during the interval of 0–10 years. Congenital anomalies showed a log–log mortality experience that was linear until the age of 45, in some populations. This linear log–log dependence corresponds to the two-parametrical Weibull distribution, if the slope is greater than −1. However, the logarithm of mortality from congenital anomalies declined with the slope equal to −1. In those congenital anomalies with the slope equal to −1, the logarithm of number of deaths in 1 year declined with the logarithm of age (that quantity corresponds to the density of defects in the theory of reliability). Consequently, there was applied another distribution function described in this paper. The number of deaths due to congenital anomalies over the period of 1 year was inversely proportional to the age at death. A hypothesis that the individual has his/her span of life strictly determined by the level of malformation at the moment of birth might explain the fact. A natural consequence of this hypothesis together with the facts mentioned above is as follows: the more serious the defect was, the more frequent this defect was. © 1998 Elsevier Science Ireland Ltd. All rights reserved.