Make a single-year life table from numbers of events and at risk
make_lt.Rddescription
Based on the output from convert_survfit2tibble(), make a tibble containing
qx (probabilities of dying before reaching age x+1) by year.
Arguments
- data
A tibble produced by
convert_survfit2tibble().
Value
A tibble with four columns, the first being year. The second column,
py, gives the person-years spent in year x.The third column, mx, gives
the mortality rate in year x; the last, qx, the probability of dying
between years x and x+1.
Details
The person-years (py) lived in each year are calculated as the mean of the
cohort at age x and age x+1, i.e., deaths are assumed to occur,
on average, mid-way through the single-year interval.
The mortality rate, mx, is calculated as the number of
events (deaths) per person-year.
The probability of death between ages x and x+1 is calculated as
\(q_{x} = 1 - \exp{-n*m_{x}}\), where n=1 as intervals are single-year.
This implies the assumption of constant mortality in the 1-year interval.