Difference between revisions of "Life table"
(a few WHO links) |
(life table equations added) |
||
| Line 1: | Line 1: | ||
| + | {{method}} | ||
| + | [[Category:Health effects]] | ||
'''Life table''' method is for estimating mortality of a population in time. | '''Life table''' method is for estimating mortality of a population in time. | ||
| + | |||
| + | ==Scope== | ||
| + | |||
| + | ==Definition== | ||
| + | |||
| + | ==Procedure== | ||
| + | |||
| + | Often the [[exposure-response relationship]]s are estimated from log-linear models: | ||
| + | |||
| + | ln p(x) = α + β x, | ||
| + | |||
| + | where p(x) is the probability of event at exposure level x, exp(α) is the background risk, and β is the slope for the exposure-response function ([[ERF]]). ERF is assumed to be exponential. | ||
| + | |||
| + | Relative risk (RR) between two exposure levels x<sub>0</sub> and x is | ||
| + | |||
| + | RR = p(x)/p(x<sub>0</sub>). | ||
| + | |||
| + | Therefore, | ||
| + | |||
| + | ln(RR) = α + β x - (α + β x<sub>0</sub>) <=> β = ln(RR)/(x - x<sub>0</sub>) | ||
| + | |||
| + | The life table is a table where the a) survival of and b) the years lived by a population are followed over the lifetime of the individuals in the population. | ||
| + | |||
| + | Assuming a constant rate of mortality (k) for a given time period, the survival S is | ||
| + | |||
| + | S = S<sub>0</sub> exp(-kt), | ||
| + | |||
| + | where t is the length of the time period. If we look at the survival conditional that the population is alive in the beginning, S<sub>0</sub> (survival in the beginning) equals 1. | ||
| + | |||
| + | As we previously showed, the mortality rate can be expressed as k = exp(α + β x), and thus the survival over time period i with constant rate is | ||
| + | |||
| + | S<sub>i</sub> = S<sub>0</sub> exp( -exp(α<sub>i</sub> + β x) t<sub>i</sub>), | ||
| + | |||
| + | and the survival over several consequent time periods is (assuming that changes in exposure are reflected in mortality with only a small delay). | ||
| + | |||
| + | S = Π<sub>i</sub> (exp( -exp(α<sub>i</sub> + β x<sub>i</sub>) t<sub>i</sub>)). | ||
| + | |||
| + | The life years lived is | ||
| + | |||
| + | Y = Σ<sub>i</sub> (S<sub>i</sub>t<sub>i</sub> + (1-S<sub>i</sub>)(1/2 t<sub>i</sub>)), | ||
| + | |||
| + | assuming that those who died, lived half of the period on average. | ||
| + | |||
==See also== | ==See also== | ||
Revision as of 11:31, 29 October 2008
This page is a method.
The page identifier is Op_en1892 |
|---|
| Moderator:Nobody (see all) Click here to sign up. |
|
Give your opinion to the peer rating of the content of this page. |
| Upload data
|
Life table method is for estimating mortality of a population in time.
Contents
Scope
Definition
Procedure
Often the exposure-response relationships are estimated from log-linear models:
ln p(x) = α + β x,
where p(x) is the probability of event at exposure level x, exp(α) is the background risk, and β is the slope for the exposure-response function (ERF). ERF is assumed to be exponential.
Relative risk (RR) between two exposure levels x0 and x is
RR = p(x)/p(x0).
Therefore,
ln(RR) = α + β x - (α + β x0) <=> β = ln(RR)/(x - x0)
The life table is a table where the a) survival of and b) the years lived by a population are followed over the lifetime of the individuals in the population.
Assuming a constant rate of mortality (k) for a given time period, the survival S is
S = S0 exp(-kt),
where t is the length of the time period. If we look at the survival conditional that the population is alive in the beginning, S0 (survival in the beginning) equals 1.
As we previously showed, the mortality rate can be expressed as k = exp(α + β x), and thus the survival over time period i with constant rate is
Si = S0 exp( -exp(αi + β x) ti),
and the survival over several consequent time periods is (assuming that changes in exposure are reflected in mortality with only a small delay).
S = Πi (exp( -exp(αi + β xi) ti)).
The life years lived is
Y = Σi (Siti + (1-Si)(1/2 ti)),
assuming that those who died, lived half of the period on average.
