Premature Death as a Normative Concept

We take 'premature death' and 'premature mortality' to have the same meaning.

A superficial summary might say that there are few detailed discussions of the concept in epidemiology and other population health disciplines but cursory discussions in philosophy. However, in philosophical discussions, the concept is usually either considered prima facie normatively significant (e.g., [5]) or discarded as too arbitrary to be normatively significant (e.g., [6]).

We borrow and develop this concept from Alexandrova and Haybron [12].

Note that remaining expected life-years here presupposes knowledge of the individual's context's force of mortality paired with a notion of central tendency or the like. Remaining expected life-years is also known as mean residual life and will be strictly positive unless there is an age beyond which no-one survives. Note further that one could also employ e.g., median residual life, or other measures of averages or central tendency.

Some cause-specific and accidental mortality statistics also seems to appeal to other dimensions, such as the distinction between natural and non-natural death and internal and external causes of death.

Conventions could diffuse such disputes, though they would arguably do so without providing rationally compelling reasons for choosing any particular ABT. Arbitrary thresholding is abundant in science. Individual instances of the practice are not obviously illegitimate or inferior, even if alternatives that do not rely on arbitrary thresholds exist. This seems to hold for quantitative studies of social phenomena, which often involve operationalizations of Ballung concepts. We suspect that the charge of arbitrariness might lose some of its appeal (and rightfully so) if appropriate bodies managed to develop acceptable standards or guidelines for how life expectancy statistics should inform the practice of setting ABTs. However, in the case of premature mortality, this is easier said than done: researchers have suggested conventions for setting ABTs since, at least, the inception of potential years of life lost (PYLL) measure in the 1940s (see, e.g., [20‐22]).

The fair innings view might be another candidate. However, it is a primarily a view about priority setting in health and, as such, it requires a philosophical defense on its own [31].

For example, Crisp's intuition about 80 years as enough would lose its appeal if, say, the global average life expectancy at birth was to significantly increase at some point in the future. Nussbaum's notion of a “life of normal length,” if it relies on life expectancy statistics (which is plausible), would lose its appeal given significant increases in lifespan inequality.

Schroeder distinguishes between aspirational and counterfactual life expectancy, which we rename as ideal and realistic life expectancy [23].

The connection between this philosophical debate and priority setting in health is explored in [39].