Thread by @snirody

1/x "Life expectancy" is an 𝐚𝐯𝐞𝐫𝐚𝐠𝐞, w/a 𝐰𝐢𝐝𝐞 𝐝𝐢𝐬𝐭𝐫𝐢𝐛𝐮𝐭𝐢𝐨𝐧 & measured at various points (birth, age 60, etc.). Survival times are not normally distributed; they follow a Gompertz-like curve (mortality rates rise exponentially with age, w/significant variation).

2/x Even if life expectancy increases by more than one year per year 𝐨𝐧 𝐚𝐯𝐞𝐫𝐚𝐠𝐞, this doesn’t guarantee LEV for individuals or the population as a whole. Some individuals won’t experience this benefit due to unequal access, biological variability, other risks.

3/x Mortality rates at advanced ages rise steeply (Gompertz law) - halving mortality at 80 might add only a few years to life expectancy. Interventions must flatten or shift the Gompertz curve, since reducing early or mid-life mortality (e.g. curing cancer) shifts the curve but doesn’t eliminate the steep rise at older ages.

4/x Death rates must decrease 𝐝𝐫𝐚𝐦𝐚𝐭𝐢𝐜𝐚𝐥𝐥𝐲 𝐚𝐧𝐝 𝐜𝐨𝐧𝐭𝐢𝐧𝐮𝐨𝐮𝐬𝐥𝐲 𝐚𝐭 𝐨𝐥𝐝𝐞𝐫 𝐚𝐠𝐞𝐬 to flatten the Gompertz curve for LEV.

5/x Looking at this, I believe that the current "definition" of LEV is 𝐟𝐚𝐜𝐢𝐥𝐞 and imprecise. It needs to be much more rigorous!

What exactly? I don't know, I'm just an Everyman! The folks specializing in this area need to come up with it!

@threadreaderapp Unroll please