Paper detail

Deciphering infant mortality. Part 1: empirical evidence

This paper is not (or at least not only) about human infant mortality. In line with reliability theory, "infant" will refer here to the time interval following birth during which the mortality (or failure) rate decreases. This definition provides a systems science perspective in which birth constitutes a sudden transition which falls within the field of application of the "Transient Shock" (TS) conjecture put forward in Richmond et al. (2016c). This conjecture provides predictions about the timing and shape of the death rate peak. (i) It says that there will be a death rate spike whenever external conditions change abruptly and drastically. (ii) It predicts that after a steep rising there will be a much longer hyperbolic relaxation process. These predictions can be tested by considering living organisms for which birth is a multi-step process. Thus, for fish there are three states: egg, yolk-sac phase, young adult. The TS conjecture predicts a mortality spike at the end of the yolk-sac phase, and this timing is indeed confirmed by observation. Secondly, the hyperbolic nature of the relaxation process can be tested using high accuracy Swiss statistics which give postnatal death rates from one hour after birth up to the age of 10 years. It turns out that since the 19th century despite a great overall reduction in infant mortality, the shape of the age-specific death rate has remained basically unchanged. This hyperbolic pattern is not specific to humans. It can also be found in small primates as recorded in the archives of zoological gardens. Our ultimate objective is to set up a chain of cases which starts from simple systems and then moves up step by step to more complex organisms. The cases discussed here can be seen as initial landmarks.